-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.html
More file actions
1060 lines (992 loc) · 59.5 KB
/
index.html
File metadata and controls
1060 lines (992 loc) · 59.5 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Curating Knowledge Collections — CUNY AI Lab</title>
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link href="https://fonts.googleapis.com/css2?family=DM+Sans:ital,opsz,wght@0,9..40,300;0,9..40,400;0,9..40,500;0,9..40,600;0,9..40,700;0,9..40,800;1,9..40,400&family=JetBrains+Mono:wght@400;500&family=Fraunces:ital,opsz,wght@0,9..144,400;0,9..144,700;0,9..144,800;1,9..144,400&display=swap" rel="stylesheet">
<link rel="stylesheet" href="css/styles.css">
<link rel="stylesheet" href="css/responsive.css">
<link rel="stylesheet" href="css/animations.css">
</head>
<body>
<a href="#deck" class="skip-link">Skip to slides</a>
<!-- Logo Watermark -->
<img src="images/logo-horizontal.png" alt="" aria-hidden="true" class="logo-watermark" id="logoWatermark">
<!-- Accessible live region for slide announcements -->
<div id="slide-announce" class="sr-only" aria-live="polite" aria-atomic="true"></div>
<!-- Lightbox overlay for expanded screenshots -->
<div class="lightbox-overlay" id="lightbox" aria-hidden="true" role="dialog" aria-label="Expanded screenshot">
<button class="lightbox-close" aria-label="Close expanded view">×</button>
<img src="" alt="">
<div class="lightbox-caption"></div>
</div>
<div id="deck" role="region" aria-roledescription="slide deck" aria-label="Curating Knowledge Collections presentation">
<!-- ==========================================
SLIDE 1 — TITLE
========================================== -->
<div class="slide title-slide" role="group" aria-roledescription="slide" aria-label="Slide 1: Curating Knowledge Collections" data-slide="title" tabindex="-1">
<div class="title-slide-content">
<img src="images/logo-horizontal.png" alt="CUNY AI Lab" class="title-logo">
<h1>Curating Knowledge Collections</h1>
<p class="subtitle">A Workshop for the <a href="https://chat.ailab.gc.cuny.edu" target="_blank" rel="noopener" style="color: inherit; text-decoration: underline; text-underline-offset: 3px;">CUNY AI Lab Sandbox</a></p>
<p class="meta">March 23, 2026</p>
<p class="meta" style="margin-top: 0.5em;">Developed by Stefano Morello and Zach Muhlbauer</p>
</div>
</div>
<!-- ==========================================
SLIDE 2 — WORKSHOP ROADMAP
========================================== -->
<div class="slide layout-full-dark" role="group" aria-roledescription="slide" aria-label="Slide 2: Three Weeks, Three Skills" data-slide="roadmap" tabindex="-1">
<span class="label label-cyan">Workshop Series</span>
<h2>Three Weeks, Three Skills</h2>
<div class="roadmap">
<div class="roadmap-item" style="opacity: 0.7;">
<div class="week-label">March 16 (Last Week)</div>
<h4><a href="https://cuny-ai-lab.github.io/system-prompting" style="color: var(--accent-cyan); text-decoration: underline; text-underline-offset: 3px;">Composing System Prompts ✓</a></h4>
<p>Learned how system prompts help orchestrate and constrain AI models to address course goals</p>
</div>
<div class="roadmap-item active">
<div class="week-label">March 23 (This Week)</div>
<h4>Curating Knowledge Collections</h4>
<p>Upload syllabi, readings, and relevant sources to ground AI models in course materials</p>
</div>
<div class="roadmap-item">
<div class="week-label">March 30</div>
<h4>Customizing Skills & Tools</h4>
<p>Build specialized skills, tools, and workflows tailored to your courses</p>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 3 — SIGN IN
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 3: Sign In and Navigate to Workspace" data-slide="sign-in" tabindex="-1">
<div class="content">
<span class="label label-cyan">Getting Started</span>
<h2>Sign In and Navigate to Workspace</h2>
<p style="margin-bottom: 0.8em;">Before we begin, make sure you can access the CUNY AI Lab Sandbox.</p>
<div class="card card-cyan step-hidden" data-step="1" style="margin-bottom: 0.5em;">
<h4>1. Go to <a href="https://chat.ailab.gc.cuny.edu" target="_blank" rel="noopener" style="color: inherit; text-decoration: underline; text-underline-offset: 3px;">chat.ailab.gc.cuny.edu</a></h4>
<p>Sign in with your CUNY account. If you don’t have access yet, let us know and we’ll get you set up.</p>
</div>
<div class="card card-navy step-hidden" data-step="2" style="margin-bottom: 0.5em;">
<h4>2. Open Workspace</h4>
<p>Look for <strong>Workspace</strong> in the top left, three rows below <strong>New Chat</strong>. Click it. This is where you’ll build your model card and knowledge collection.</p>
</div>
<div class="tip-box step-hidden" data-step="3" style="margin-top: 1em;">
<p><strong>Next:</strong> Once you’re signed in and can see the Workspace menu, you’re ready to set up your model card on the next slide.</p>
</div>
</div>
<div class="stage">
<div class="carousel" data-interval="8000" data-sync-steps>
<div class="carousel-item active">
<img src="images/slide3-workspace-create.png" alt="Open WebUI sidebar showing Workspace menu item">
<div class="carousel-caption"><strong>Workspace → Models</strong><br>Navigate to Workspace in the sidebar, then open the Models tab.</div>
</div>
<div class="carousel-item">
<img src="images/slide3-model-basics.png" alt="Workspace Models page listing custom models in the Sandbox">
<div class="carousel-caption"><strong>Models List</strong><br>Click <strong>+ New Model</strong> to create one, or open an existing model card.</div>
</div>
<div class="carousel-dots"></div>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 4 — SET UP YOUR MODEL CARD
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 4: Set Up Your Model Card" data-slide="model-setup" tabindex="-1">
<div class="content">
<span class="label label-cyan">Before We Start</span>
<h2>Set Up Your Model Card</h2>
<p style="margin-bottom: 0.8em;">Knowledge collections attach to a model card in <strong>Workspace → Models</strong>. After you build your collection, complete these steps to bring everything together.</p>
<div class="card card-cyan step-hidden" data-step="1" style="margin-bottom: 0.5em;">
<h4>1. Name the model card</h4>
<p>Give it a descriptive name tied to your course, for example <em>ENGL 101 Writing Scaffold</em> or <em>History 202 Source Analysis Tool</em>.</p>
</div>
<div class="card step-hidden" data-step="2" style="margin-bottom: 0.5em;">
<h4>2. Select a base model</h4>
<p>Choose a base model from the Sandbox (e.g., GLM-5, GPT-OSS-120b, or Kimi-k2.5). This is the underlying model your custom configuration will run on.</p>
</div>
<div class="card step-hidden" data-step="3" style="margin-bottom: 0.5em;">
<h4>3. Add your system prompt</h4>
<p>Find the <strong>System Prompt</strong> field under Model Params and paste the prompt you wrote last week.</p>
</div>
<div class="card card-navy step-hidden" data-step="4" style="margin-bottom: 0.5em;">
<h4>4. Save your work</h4>
<p>Click the <strong>Save</strong> button. Your model card is now ready for a knowledge collection.</p>
</div>
</div>
<div class="stage" style="flex-direction: column; justify-content: center; padding: clamp(16px, 2.5vw, 40px);">
<div class="carousel" data-interval="8000">
<div class="carousel-item active">
<img src="images/slide3-system-prompt.png" alt="Model card editor showing fields for name, base model, system prompt, and knowledge collection">
<div class="carousel-caption"><strong>Inside the Model Card</strong><br>Name your model, select a base model, paste your system prompt, and save. The <strong>Knowledge</strong> section below is where today’s collection will attach.</div>
</div>
<div class="carousel-dots"></div>
</div>
<div class="tip-box step-hidden" data-step="5" style="margin-top: 1.2em;">
<p><strong>Keep this tab open.</strong> You will return to this model card at the end of today’s exercise to attach the knowledge collection you build.</p>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 5 — TEST YOUR MODEL IN CHAT
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 5: Test Your Model in Chat" data-slide="test-model" tabindex="-1">
<div class="content">
<span class="label label-cyan">Dry Run</span>
<h2>Test Your Model in Chat</h2>
<p style="margin-bottom: 0.8em;">Your model is saved. Now test it to make sure it responds as expected before attaching a knowledge collection.</p>
<div class="card card-cyan" style="margin-bottom: 0.5em;">
<h4>1. Click New Chat</h4>
<p>Go back to the chat interface. Click <strong>New Chat</strong> in the top left to start a fresh conversation.</p>
</div>
<div class="card" style="margin-bottom: 0.5em;">
<h4>2. Toggle the model selector</h4>
<p>At the top of the chat window, you’ll see the model name. Click the dropdown arrow next to it.</p>
</div>
<div class="card" style="margin-bottom: 0.5em;">
<h4>3. Select your model</h4>
<p>Find your newly created model (e.g., <strong>ENGL 101 Writing Scaffold</strong>) in the dropdown list and select it.</p>
</div>
<div class="card card-navy" style="margin-bottom: 0.5em;">
<h4>4. Send a test message</h4>
<p>Type a simple message related to your course. Send it and observe the response.</p>
</div>
</div>
<div class="stage" style="flex-direction: column; justify-content: center; align-items: center; padding: clamp(16px, 2.5vw, 40px);">
<img src="images/slide5.png" alt="Open WebUI chat interface showing the model selector dropdown with a custom model selected, and the message input field" class="screenshot-img">
<div class="carousel-caption"><strong>Test Your Model</strong><br>Click <strong>New Chat</strong> (1), select your model from the dropdown (2-3), and type a test message (4).</div>
<div class="tip-box" style="margin-top: 1.2em;">
<p><strong>Look for:</strong> Does your custom model respond as expected? How has your system prompt shaped its output?</p>
</div>
</div>
</div>
<!-- ==========================================
SECTION DIVIDER — KNOWLEDGE COLLECTIONS
========================================== -->
<div class="slide layout-divider" role="group" aria-roledescription="slide" aria-label="Slide 6: Knowledge Collections" data-slide="kc-intro" tabindex="-1">
<span class="section-label">Part I</span>
<h2>Knowledge Collections</h2>
<p class="divider-sub" style="margin-top: 0.5em; opacity: 0.7; font-size: 0.95rem;">What they are, where they live, and how they work</p>
<div class="divider-line"></div>
</div>
<!-- ==========================================
SLIDE — WHAT IS A KNOWLEDGE COLLECTION?
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 7: What Is a Knowledge Collection?" data-slide="what-is" tabindex="-1">
<div class="content" style="flex: 0 0 42%;">
<span class="label label-cyan">The Basics</span>
<h2>What Is a Knowledge Collection?</h2>
<p style="margin-bottom: 0.8em;">A <strong>knowledge collection</strong> is a set of documents you upload to ground an AI model in your course materials. The model retrieves relevant passages from these documents when responding.</p>
<blockquote>Think of it as a <strong>reference shelf</strong> for your custom AI model: the sources it draws on, the assignments it references, and the disciplinary context that shapes its responses.</blockquote>
<p style="margin-top: 1em;"><em>Next we’ll show you where to find knowledge collections in the Sandbox and how to curate and use them with pedagogical intent.</em></p>
</div>
<div class="stage" style="background: var(--bg-light); flex-direction: column; padding: clamp(16px, 2.5vh, 32px) clamp(14px, 1.8vw, 24px); justify-content: center;">
<div class="theater-diagram">
<!-- Behind the curtain: instructor builds both -->
<div class="theater-backstage">
<div class="theater-zone-label">What you build</div>
<div class="theater-backstage-row">
<div class="flow-node flow-prompt" style="width: auto; flex: 1;">
<span class="flow-hidden-badge">Hidden</span>
<div class="flow-node-code" style="font-size: 0.85rem;">System Prompt</div>
<div class="flow-node-desc" style="font-family: 'JetBrains Mono', monospace; color: var(--accent-cyan); margin-top: 3px; font-size: 0.72rem;">“You are a writing scaffold…”</div>
</div>
<div style="display: flex; align-items: center; padding: 0 6px; font-size: 0.9rem; color: var(--accent-cyan); font-weight: 700;">+</div>
<div class="flow-node flow-prompt" style="width: auto; flex: 1; border-color: rgba(59, 111, 181, 0.4);">
<span class="flow-hidden-badge" style="background: var(--accent-blue);">New</span>
<div class="flow-node-code" style="font-size: 0.85rem;">Knowledge Collection</div>
<div class="flow-node-desc" style="font-family: 'JetBrains Mono', monospace; color: var(--accent-light-blue); margin-top: 3px; font-size: 0.72rem;">syllabus.pdf, prompt.pdf…</div>
</div>
</div>
</div>
<!-- The curtain line -->
<div class="theater-curtain">
<div class="theater-curtain-line"></div>
<span class="theater-curtain-label">student never sees above</span>
<div class="theater-curtain-line"></div>
</div>
<!-- Front of house: student interaction -->
<div class="theater-frontstage">
<div class="theater-conversation">
<div class="theater-msg theater-msg-user">
<span class="theater-msg-who"><span aria-hidden="true">👤</span> Student</span>
<span class="theater-msg-text">“What does the assignment say about evidence?”</span>
</div>
</div>
<!-- AI Model sits between user prompt and response -->
<div class="flow-node flow-model" style="width: 100%; margin: clamp(6px, 1vh, 10px) 0;">
<div class="flow-node-icon" style="font-size: 1.5rem;" aria-hidden="true">🤖</div>
<div class="flow-node-label" style="font-size: 0.92rem;">AI Model</div>
<div class="flow-node-desc" style="font-size: 0.76rem;">Retrieves passages from uploaded course materials</div>
</div>
<div class="theater-conversation">
<div class="theater-msg theater-msg-ai">
<span class="theater-msg-who"><span aria-hidden="true">🤖</span> Response</span>
<span class="theater-msg-text">“The assignment asks for at least three sources cited with specific page numbers. Let’s look at what you have so far.”</span>
</div>
</div>
<div class="theater-zone-label" style="color: var(--accent-blue); margin-top: 6px;">What the student sees</div>
</div>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 4 — WHERE IT LIVES IN OPEN WEBUI
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 7: Where to Find Knowledge Collections" data-slide="where-it-lives" tabindex="-1">
<div class="content">
<span class="label label-cyan">Workspace</span>
<h2>Where to Find Knowledge Collections</h2>
<p style="margin-bottom: 0.8em;">Knowledge collections and model cards both live in the <strong>Workspace</strong> menu. Here’s how they connect:</p>
<div class="card card-cyan" style="margin-bottom: 0.6em;">
<h4>Workspace → Knowledge</h4>
<p>Where collections are created, named, and populated with course documents.</p>
</div>
<div class="card" style="margin-bottom: 0.6em;">
<h4>Workspace → Models</h4>
<p>Where collections are attached to a model card via the <strong>Knowledge</strong> field, giving the model access to your materials.</p>
</div>
<div class="tip-box">
<p><strong>Key distinction:</strong> Your system prompt defines how the model responds. The knowledge collection provides the source material it draws on. Both are configured in the same model card.</p>
</div>
</div>
<div class="stage">
<div class="carousel" data-interval="8000">
<div class="carousel-item active">
<img src="images/slide4-a.png" alt="Workspace Knowledge tab showing existing collections and the New Knowledge button">
<div class="carousel-caption"><strong>Workspace → Knowledge</strong><br>Click <strong>+ New Knowledge</strong> to create a collection.</div>
</div>
<div class="carousel-item">
<img src="images/slide4-b.png" alt="Create a knowledge base form with name, description, and privacy settings">
<div class="carousel-caption"><strong>Name and Describe</strong><br>Give your collection a name, describe its contents, and set access. Click <strong>Create Knowledge</strong> when ready.</div>
</div>
<div class="carousel-item">
<img src="images/slide4-c.png" alt="Workspace Models page listing custom models">
<div class="carousel-caption"><strong>Workspace → Models</strong><br>Open your model card to attach the collection.</div>
</div>
<div class="carousel-item">
<img src="images/slide4-d.png" alt="Model card editor with Knowledge section highlighted showing Select Knowledge and Upload Files">
<div class="carousel-caption"><strong>Knowledge Field</strong><br>Click <strong>Select Knowledge</strong> to attach your collection to the model card.</div>
</div>
<div class="carousel-dots"></div>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 5 — WHY IT MATTERS
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 8: Why Knowledge Collections Matter" data-slide="why-it-matters" tabindex="-1">
<span class="label label-cyan">Why Bother?</span>
<h2>Why Knowledge Collections Matter</h2>
<p style="margin-bottom: 0.6em;">If you create a custom model that acts as a course assistant, a student might ask: <strong>“What should I focus on for the midterm essay?”</strong></p>
<div class="comparison" data-steps="2">
<div class="comparison-panel panel-muted step-visible">
<h4 style="color: var(--accent-red);">Without Knowledge Collection</h4>
<p>“For a midterm essay, you should generally focus on your thesis statement, use evidence from your readings, and structure your argument clearly. Make sure to address counterarguments.”</p>
<p style="margin-top: 0.6em; font-style: italic; font-size: 0.85em; opacity: 0.7;">Generic advice. No connection to the assignment or the course readings.</p>
</div>
<div class="comparison-panel step-hidden" data-step="2">
<h4 style="color: var(--accent-green);">With Knowledge Collection</h4>
<p>“Based on the assignment prompt, your essay should analyze one primary source from the Reconstruction unit using the SOAPS framework we practiced. The prompt emphasizes evidence and sourcing. Which document are you considering?”</p>
<p style="margin-top: 0.6em; font-style: italic; font-size: 0.85em; opacity: 0.7;">Grounded in the actual assignment and course methodology.</p>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 6 — WHAT CAN YOU UPLOAD?
========================================== -->
<div class="slide layout-full-dark" role="group" aria-roledescription="slide" aria-label="Slide 9: What Can You Upload?" data-slide="upload-types" tabindex="-1">
<span class="label label-cyan">Building Materials</span>
<h2>What Can You Upload?</h2>
<p style="color: var(--light-text-50); max-width: 700px; margin-bottom: 1.2em;">Uploaded files ground AI models in context and help shape their responses. These documents are stored on CUNY’s self-hosted servers and made private by default.</p>
<div class="model-grid">
<div class="model-card">
<h4>Syllabi</h4>
<p>Course schedule, learning objectives, policies, and expectations</p>
</div>
<div class="model-card">
<h4>Assignment Prompts</h4>
<p>Instructions, requirements, and criteria for each assignment</p>
</div>
<div class="model-card">
<h4>Rubrics</h4>
<p>Evaluation criteria so the model can reference specific expectations</p>
</div>
<div class="model-card" style="border-left-color: var(--accent-cyan);">
<h4>Course Readings</h4>
<p>Primary sources, articles, chapters, and excerpts students are working with</p>
</div>
<div class="model-card" style="border-left-color: var(--accent-cyan);">
<h4>Lecture Notes</h4>
<p>Key concepts, frameworks, and terminology from your lectures</p>
</div>
<div class="model-card" style="border-left-color: var(--accent-light-blue);">
<h4>Style Guides</h4>
<p>Citation formats, disciplinary conventions, writing guidelines</p>
</div>
<div class="model-card" style="border-left-color: var(--accent-light-blue);">
<h4>Sample Work</h4>
<p>Exemplars that model the kinds of work you expect</p>
</div>
<div class="model-card" style="border-left-color: var(--accent-amber);">
<h4>Data Sets</h4>
<p>Spreadsheets, CSV files, or structured data students analyze in labs or projects</p>
</div>
<div class="model-card" style="border-left-color: var(--accent-amber);">
<h4>Glossaries</h4>
<p>Discipline-specific terminology, definitions, and key concepts for the course</p>
</div>
<div class="model-card" style="border-left-color: var(--accent-amber);">
<h4>Problem Sets</h4>
<p>Exercises, practice questions, or worked examples with solutions</p>
</div>
<div class="model-card" style="border-left-color: var(--accent-green);">
<h4>Lab Protocols</h4>
<p>Step-by-step procedures, safety guidelines, and equipment instructions</p>
</div>
<div class="model-card" style="border-left-color: var(--accent-green);">
<h4>Case Studies</h4>
<p>Real-world scenarios, historical cases, or clinical examples used in coursework</p>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 7 — RETRIEVAL (merged: was slides 7+8)
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 10: How the Model Uses Your Documents" data-slide="retrieval" tabindex="-1">
<div class="content">
<span class="section-label" style="font-size: 0.75rem; letter-spacing: 0.08em; text-transform: uppercase; color: var(--accent-cyan); margin-bottom: 0.3em; display: block;">Under the Hood</span>
<span class="label label-cyan">Retrieval-Augmented Generation</span>
<h2>How Your Documents Reach the Model</h2>
<p style="margin-bottom: 0.8em;">When a student asks a question, the system doesn’t feed the <em>entire</em> collection to the model. It searches for the most relevant passages and uses them as the basis for its response.</p>
<ul style="margin-top: 0.8em; line-height: 1.9;">
<li><strong>Chunking:</strong> Your files are split into smaller passages when uploaded</li>
<li><strong>Matching:</strong> Questions are matched against those chunks</li>
<li><strong>Injection:</strong> Closest matches appended to model’s context window</li>
<li><strong>Response:</strong> Model generates output grounded in retrieved passages</li>
</ul>
<div class="tip-box" style="margin-top: 0.8em;">
<p><strong>Implication:</strong> Short, focused documents with clear headings retrieve better than long, unstructured files. In other words, the way you organize your materials matters.</p>
</div>
</div>
<div class="stage" style="background: var(--bg-light); flex-direction: column; padding: clamp(16px, 2.5vw, 32px); justify-content: center;">
<div style="display: flex; flex-direction: column; gap: clamp(10px, 1.5vh, 18px); width: 100%; max-width: 420px; margin: 0 auto;">
<!-- Step 1: Student question -->
<div class="flow-node flow-student" style="width: 100%;">
<div class="flow-node-icon" style="font-size: 1.5rem;" aria-hidden="true">👤</div>
<div class="flow-node-label" style="font-size: 0.95rem;">Student Question</div>
<div class="flow-node-desc" style="font-size: 0.8rem;">“What does the assignment say about citations?”</div>
</div>
<!-- Arrow down -->
<div style="text-align: center;" aria-hidden="true">
<svg width="24" height="24" viewBox="0 0 20 20" focusable="false"><path d="M10 2 L10 14 M6 10 L10 16 L14 10" stroke="var(--accent-cyan)" stroke-width="2" fill="none" stroke-linecap="round"/></svg>
</div>
<!-- Step 2: Search -->
<div class="flow-node" style="width: 100%; background: rgba(78, 205, 196, 0.08); border: 2px solid rgba(78, 205, 196, 0.3); border-radius: 12px;">
<div class="flow-node-icon" style="font-size: 1.5rem;" aria-hidden="true">🔍</div>
<div class="flow-node-label" style="font-size: 0.95rem;">Search Collection</div>
<div class="flow-node-desc" style="font-size: 0.8rem;">Finds relevant passages from your files</div>
</div>
<!-- Arrow down -->
<div style="text-align: center;" aria-hidden="true">
<svg width="24" height="24" viewBox="0 0 20 20" focusable="false"><path d="M10 2 L10 14 M6 10 L10 16 L14 10" stroke="var(--accent-blue)" stroke-width="2" fill="none" stroke-linecap="round"/></svg>
</div>
<!-- Step 3: Model + context -->
<div class="flow-node flow-model" style="width: 100%;">
<div class="flow-node-icon" style="font-size: 1.5rem;" aria-hidden="true">🤖</div>
<div class="flow-node-label" style="font-size: 0.95rem;">AI Model + Retrieved Context</div>
<div class="flow-node-desc" style="font-size: 0.8rem;">System prompt + relevant passages + student message</div>
</div>
<!-- Arrow down -->
<div style="text-align: center;" aria-hidden="true">
<svg width="24" height="24" viewBox="0 0 20 20" focusable="false"><path d="M10 2 L10 14 M6 10 L10 16 L14 10" stroke="var(--accent-green)" stroke-width="2" fill="none" stroke-linecap="round"/></svg>
</div>
<!-- Step 4: Grounded response -->
<div class="flow-node" style="width: 100%; background: white; border: 2px solid var(--accent-green); border-radius: 12px; box-shadow: 0 4px 20px rgba(91, 138, 111, 0.1);">
<div class="flow-node-icon" style="font-size: 1.5rem;" aria-hidden="true">✅</div>
<div class="flow-node-label" style="font-size: 0.95rem;">Grounded Response</div>
<div class="flow-node-desc" style="font-size: 0.8rem;">Answer references your actual course materials</div>
</div>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 8 — SECTION DIVIDER: EXAMPLES (merged: was slides 9+10)
========================================== -->
<div class="slide layout-divider" role="group" aria-roledescription="slide" aria-label="Slide 11: What Makes an Effective Knowledge Collection?" data-slide="part1" tabindex="-1">
<span class="section-label">Part II: Example 1</span>
<h2>What Makes an Effective Knowledge Collection?</h2>
<p class="divider-sub">Starting with Composition & Writing</p>
<div class="divider-line"></div>
</div>
<!-- ==========================================
SLIDE 11 — COMPOSITION: BAD COLLECTION
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 12: Composition: The Bare Minimum" data-slide="composition-bad" tabindex="-1">
<div class="progression-header">
<span class="label label-blue" style="margin-bottom: 0;">Composition & Writing</span>
<div class="progression-dots" aria-hidden="true">
<div class="progression-dot dot-bad dot-active"></div>
<div class="progression-dot"></div>
<div class="progression-dot"></div>
</div>
</div>
<h2>The Bare Minimum</h2>
<div class="prompt-block prompt-bad"><span class="prompt-label">Weak</span>Collection contents:
• syllabus.pdf (14 pages, full course syllabus)</div>
<div style="margin-top: 1em;">
<h4 style="color: var(--accent-red);">What goes wrong?</h4>
<ul class="stream-list step-hidden" data-step>
<li>One large document retrieves poorly: retrieved passages are often irrelevant</li>
<li>No assignment context for the revision task</li>
<li>No readings or reference materials for the model to draw on</li>
</ul>
</div>
</div>
<!-- ==========================================
SLIDE 12 — COMPOSITION: MEDIOCRE COLLECTION
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 13: Composition: Getting Warmer" data-slide="composition-mid" tabindex="-1">
<div class="progression-header">
<span class="label label-blue" style="margin-bottom: 0;">Composition & Writing</span>
<div class="progression-dots" aria-hidden="true">
<div class="progression-dot dot-bad"></div>
<div class="progression-dot dot-mid dot-active"></div>
<div class="progression-dot"></div>
</div>
</div>
<h2>Getting Warmer</h2>
<div class="prompt-block prompt-mid"><span class="prompt-label">Getting There</span>Collection contents:
• syllabus.pdf
• essay-1-prompt.pdf
• mla-style-guide.pdf</div>
<div class="why-better">
<h4>What improved?</h4>
<ul class="stream-list step-hidden" data-step>
<li>Separate documents let the model find what it needs</li>
<li>Assignment prompt gives the model context for the revision task</li>
<li>Style guide helps with formatting questions</li>
</ul>
</div>
<div style="margin-top: 0.6em;">
<h4 style="color: var(--accent-amber);">What's still missing?</h4>
<ul class="stream-list step-hidden" data-step>
<li>No course readings for the model to reference during analysis</li>
<li>No common feedback patterns to guide revision</li>
<li>No instructor notes on what substantive revision looks like in this course</li>
</ul>
</div>
</div>
<!-- ==========================================
SLIDE 13 — COMPOSITION: GREAT COLLECTION
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 14: A Collection That Grounds Revision" data-slide="composition-good" tabindex="-1">
<div class="progression-header">
<span class="label label-blue" style="margin-bottom: 0;">Composition & Writing</span>
<div class="progression-dots" aria-hidden="true">
<div class="progression-dot dot-bad"></div>
<div class="progression-dot dot-mid"></div>
<div class="progression-dot dot-good dot-active"></div>
</div>
</div>
<h2>A Collection That Grounds Revision</h2>
<div class="prompt-block prompt-good" style="font-size: clamp(0.85rem, 1.15vw, 1rem);"><span class="prompt-label">Strong</span>Collection contents:
<strong style="color: var(--accent-cyan);">Course Context</strong>
• syllabus.pdf: schedule, learning objectives, policies
• revision-philosophy.txt: instructor notes on what revision means in this course
<strong style="color: var(--accent-cyan);">Assignment Materials (Essay 1: Rhetoric in Popular Media)</strong>
• essay-1-prompt.pdf: assignment instructions and requirements
• common-feedback.txt: patterns from past semesters (e.g., thesis too broad, evidence not analyzed)
<strong style="color: var(--accent-cyan);">Reference Materials</strong>
• mla-style-guide.pdf: citation and formatting conventions
• strong-intro-examples.txt: examples of effective introductions
• revision-checklist.pdf: the same checklist students use in peer review</div>
</div>
<!-- ==========================================
SLIDE 14 — SECTION DIVIDER: HISTORY
========================================== -->
<div class="slide layout-divider" role="group" aria-roledescription="slide" aria-label="Slide 15: Primary Source Analysis" data-slide="history-intro" tabindex="-1">
<span class="section-label">Example 2</span>
<h2>Primary Source Analysis</h2>
<div class="divider-line"></div>
</div>
<!-- ==========================================
SLIDE 15 — HISTORY: BAD COLLECTION
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 16: History: The Bare Minimum" data-slide="history-bad" tabindex="-1">
<div class="progression-header">
<span class="label label-blue" style="margin-bottom: 0;">History</span>
<div class="progression-dots" aria-hidden="true">
<div class="progression-dot dot-bad dot-active"></div>
<div class="progression-dot"></div>
<div class="progression-dot"></div>
</div>
</div>
<h2>The Bare Minimum</h2>
<div class="prompt-block prompt-bad"><span class="prompt-label">Weak</span>Collection contents:
• textbook-chapter-12.pdf (42 pages)</div>
<div style="margin-top: 1em;">
<h4 style="color: var(--accent-red);">What goes wrong?</h4>
<ul class="stream-list step-hidden" data-step>
<li>A full textbook chapter is too long and too general: retrieved passages are often irrelevant</li>
<li>No primary sources for the model to help students analyze</li>
<li>No framework like SOAPS for the model to scaffold source analysis</li>
</ul>
</div>
</div>
<!-- ==========================================
SLIDE 16 — HISTORY: MEDIOCRE COLLECTION
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 17: History: Getting Warmer" data-slide="history-mid" tabindex="-1">
<div class="progression-header">
<span class="label label-blue" style="margin-bottom: 0;">History</span>
<div class="progression-dots" aria-hidden="true">
<div class="progression-dot dot-bad"></div>
<div class="progression-dot dot-mid dot-active"></div>
<div class="progression-dot"></div>
</div>
</div>
<h2>Getting Warmer</h2>
<div class="prompt-block prompt-mid"><span class="prompt-label">Getting There</span>Collection contents:
• syllabus.pdf
• source-analysis-assignment.pdf
• primary-source-1.pdf (Freedmen's Bureau report, 1866)
• primary-source-2.pdf (Congressional testimony, 1871)</div>
<div class="why-better">
<h4>What improved?</h4>
<ul class="stream-list step-hidden" data-step>
<li>Includes actual primary sources students are working with</li>
<li>Assignment prompt gives the model task-specific context</li>
<li>Documents are separate and focused</li>
</ul>
</div>
<div style="margin-top: 0.6em;">
<h4 style="color: var(--accent-amber);">What's still missing?</h4>
<ul class="stream-list step-hidden" data-step>
<li>No contextual background for the model to draw on when students ask about the period</li>
<li>No SOAPS framework or equivalent to guide source analysis</li>
<li>No source metadata (author, date, document type) to support sourcing questions</li>
</ul>
</div>
</div>
<!-- ==========================================
SLIDE 17 — HISTORY: GREAT COLLECTION
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 18: A Collection That Fosters Historical Thinking" data-slide="history-good" tabindex="-1">
<div class="progression-header">
<span class="label label-blue" style="margin-bottom: 0;">History</span>
<div class="progression-dots" aria-hidden="true">
<div class="progression-dot dot-bad"></div>
<div class="progression-dot dot-mid"></div>
<div class="progression-dot dot-good dot-active"></div>
</div>
</div>
<h2>A Collection That Fosters Historical Thinking</h2>
<div class="prompt-block prompt-good" style="font-size: clamp(0.85rem, 1.15vw, 1rem);"><span class="prompt-label">Strong</span>Collection contents:
<strong style="color: var(--accent-cyan);">Course Context</strong>
• syllabus.pdf: schedule, themes, learning objectives
• soaps-framework.txt: the analytical framework students use, with definitions and examples
<strong style="color: var(--accent-cyan);">Primary Sources (Reconstruction Unit)</strong>
• freedmens-bureau-report-1866.pdf: with metadata: author, date, document type, archive
• congressional-testimony-1871.pdf: with metadata
• source-context-notes.txt: brief historical context for each source (2-3 sentences each)
<strong style="color: var(--accent-cyan);">Reference Materials</strong>
• period-timeline.txt: key events 1865-1877 for contextualization questions
• common-analysis-errors.txt: patterns from past semesters (e.g., treating sources as neutral facts)
• chicago-citation-guide.pdf: citation format for history papers</div>
</div>
<!-- ==========================================
SLIDE 18 — SECTION DIVIDER: LITERATURE
========================================== -->
<div class="slide layout-divider" role="group" aria-roledescription="slide" aria-label="Slide 19: Close Reading and Literary Analysis" data-slide="literature-intro" tabindex="-1">
<span class="section-label">Example 3</span>
<h2>Close Reading & Literary Analysis</h2>
<div class="divider-line"></div>
</div>
<!-- ==========================================
SLIDE 19 — LITERATURE: BAD COLLECTION
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 20: Literature: The Bare Minimum" data-slide="literature-bad" tabindex="-1">
<div class="progression-header">
<span class="label label-blue" style="margin-bottom: 0;">Literature & Cultural Studies</span>
<div class="progression-dots" aria-hidden="true">
<div class="progression-dot dot-bad dot-active"></div>
<div class="progression-dot"></div>
<div class="progression-dot"></div>
</div>
</div>
<h2>The Bare Minimum</h2>
<div class="prompt-block prompt-bad"><span class="prompt-label">Weak</span>Collection contents:
• course-reader.pdf (180 pages, all readings for the semester)</div>
<div style="margin-top: 1em;">
<h4 style="color: var(--accent-red);">What goes wrong?</h4>
<ul class="stream-list step-hidden" data-step>
<li>A 180-page file retrieves unpredictably: the model might pull from the wrong text entirely</li>
<li>No assignment context or close-reading framework</li>
<li>No separation between literary texts and critical essays</li>
</ul>
</div>
</div>
<!-- ==========================================
SLIDE 20 — LITERATURE: MEDIOCRE COLLECTION
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 21: Literature: Getting Warmer" data-slide="literature-mid" tabindex="-1">
<div class="progression-header">
<span class="label label-blue" style="margin-bottom: 0;">Literature & Cultural Studies</span>
<div class="progression-dots" aria-hidden="true">
<div class="progression-dot dot-bad"></div>
<div class="progression-dot dot-mid dot-active"></div>
<div class="progression-dot"></div>
</div>
</div>
<h2>Getting Warmer</h2>
<div class="prompt-block prompt-mid"><span class="prompt-label">Getting There</span>Collection contents:
• syllabus.pdf
• close-reading-assignment.pdf
• sonny-blues-baldwin.pdf
• new-criticism-overview.pdf</div>
<div class="why-better">
<h4>What improved?</h4>
<ul class="stream-list step-hidden" data-step>
<li>Individual literary text rather than an omnibus reader</li>
<li>Assignment prompt provides task-specific context</li>
<li>Critical framework document gives the model methodological grounding</li>
</ul>
</div>
<div style="margin-top: 0.6em;">
<h4 style="color: var(--accent-amber);">What's still missing?</h4>
<ul class="stream-list step-hidden" data-step>
<li>No annotated examples showing how to move from observation to interpretation</li>
<li>No key terms for the current unit (e.g., tension, irony, ambiguity)</li>
<li>No instructor notes on what close reading looks like in this course</li>
</ul>
</div>
</div>
<!-- ==========================================
SLIDE 21 — LITERATURE: GREAT COLLECTION
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 22: A Collection That Fosters Close Reading" data-slide="literature-good" tabindex="-1">
<div class="progression-header">
<span class="label label-blue" style="margin-bottom: 0;">Literature & Cultural Studies</span>
<div class="progression-dots" aria-hidden="true">
<div class="progression-dot dot-bad"></div>
<div class="progression-dot dot-mid"></div>
<div class="progression-dot dot-good dot-active"></div>
</div>
</div>
<h2>A Collection That Fosters Close Reading</h2>
<div class="prompt-block prompt-good" style="font-size: clamp(0.85rem, 1.15vw, 1rem);"><span class="prompt-label">Strong</span>Collection contents:
<strong style="color: var(--accent-cyan);">Course Context</strong>
• syllabus.pdf: schedule, texts, learning objectives
• new-criticism-framework.txt: key concepts and terms for this unit (tension, irony, paradox, ambiguity, diction, imagery)
<strong style="color: var(--accent-cyan);">Assignment Materials (Close Reading Essay)</strong>
• close-reading-assignment.pdf: instructions and requirements
• annotated-passage-example.txt: model annotation showing how to move from observation to interpretation
<strong style="color: var(--accent-cyan);">Literary Texts (Current Unit)</strong>
• sonny-blues-baldwin.pdf: the primary text for this assignment
• passage-selections.txt: key passages the instructor has flagged for class discussion</div>
</div>
<!-- ==========================================
SLIDE 23 — SECTION DIVIDER: BEST PRACTICES
========================================== -->
<div class="slide layout-divider" role="group" aria-roledescription="slide" aria-label="Slide 23: Curating for Retrieval Quality" data-slide="part3" tabindex="-1">
<span class="section-label">Part III</span>
<h2>Curating for<br>Retrieval Quality</h2>
<p class="divider-sub">Tips for curating collections, and common pitfalls to avoid</p>
<div class="divider-line"></div>
</div>
<!-- ==========================================
SLIDE 24 — BEST PRACTICES
========================================== -->
<div class="slide layout-grid" role="group" aria-roledescription="slide" aria-label="Slide 24: Curation Best Practices" data-slide="best-practices" tabindex="-1">
<h2>Curation Best Practices</h2>
<div class="grid-2">
<div class="card card-cyan">
<h4>One Document, One Purpose</h4>
<p style="font-size: 1.15em;">Upload separate files; focused documents retrieve better than omnibus ones</p>
</div>
<div class="card">
<h4>Add Metadata and Headings</h4>
<p style="font-size: 1.15em;">Titles, authors, dates, and section headings serve as retrieval anchors</p>
</div>
<div class="card card-navy">
<h4>Supply What’s Not in the Documents</h4>
<p style="font-size: 1.15em;">Include meta documents like “common-feedback.txt” that signpost how to use sources in the collection</p>
</div>
<div class="card card-cyan">
<h4>Update Per Unit</h4>
<p style="font-size: 1.15em;">Swap course materials as the semester progresses; up-to-date collections outperform semester-wide ones</p>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 25 — COMMON PITFALLS
========================================== -->
<div class="slide layout-content" role="group" aria-roledescription="slide" aria-label="Slide 25: Common Pitfalls" data-slide="pitfalls" tabindex="-1">
<span class="label label-cyan">Watch Out</span>
<h2>Common Pitfalls</h2>
<div class="grid-2" style="margin-top: 1em;">
<div class="card" style="border-left-color: var(--accent-red);">
<h4>Dumping Everything In</h4>
<p style="font-size: 1.15em;">Uploading every reading dilutes retrieval; start small and add materials as you test</p>
</div>
<div class="card" style="border-left-color: var(--accent-red);">
<h4>One Giant PDF</h4>
<p style="font-size: 1.15em;">A 200-page course reader retrieves unpredictably; short, well-labeled documents work far better</p>
</div>
<div class="card" style="border-left-color: var(--accent-amber);">
<h4>Forgetting the System Prompt</h4>
<p style="font-size: 1.15em;">Without explicit instructions for drawing on the collection, it is just a pile of documents</p>
</div>
<div class="card" style="border-left-color: var(--accent-amber);">
<h4>Assuming Full Coverage</h4>
<p style="font-size: 1.15em;">Only retrieved passages appear in each response; if something is critical, give it its own file</p>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 26 — SECTION DIVIDER: HANDS-ON EXERCISE
========================================== -->
<div class="slide layout-divider" role="group" aria-roledescription="slide" aria-label="Slide 26: Building Your Knowledge Collection" data-slide="part4" tabindex="-1">
<span class="section-label">Part IV</span>
<h2>Building Your<br>Knowledge Collection</h2>
<p class="divider-sub">Three types of references to consider, then steps for how to create, curate, and use your first collection.</p>
<div class="divider-line"></div>
</div>
<!-- ==========================================
SLIDE 27 — THREE TYPES OF REFERENCE MATERIAL
========================================== -->
<div class="slide layout-content" style="justify-content: center; align-items: center; text-align: center;" role="group" aria-roledescription="slide" aria-label="Slide 27: Types of Reference Material" data-slide="three-layers" tabindex="-1">
<h2>Types of Reference Material</h2>
<p style="margin-bottom: 1em; max-width: 600px; margin-left: auto; margin-right: auto;">Think about which type of course document you would add first</p>
<ol class="stream-list" style="text-align: left; max-width: 580px; margin: 0 auto; line-height: 1.9; padding-left: 1.4em;">
<li><strong>Course Context:</strong> Syllabus sections, weekly schedule</li>
<li><strong>Assignment Materials:</strong> Instructions, feedback examples</li>
<li><strong>Source Materials:</strong> Excerpted readings, primary sources</li>
</ol>
</div>
<!-- ==========================================
SLIDE 28 — LAYER 1: COURSE FRAMEWORK
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 28: Course Context" data-slide="layer-framework" tabindex="-1">
<div class="content">
<span class="label label-cyan">Type 1</span>
<h2>Course Context</h2>
<p>These documents give the model a picture of your course: its goals, structure, and the methods students are expected to use.</p>
<ul style="margin-top: 1em; line-height: 1.9;">
<li>What are the course’s learning objectives?</li>
<li>What analytical framework or methodology is central to the course?</li>
<li>What course-level context would help the model support those goals?</li>
</ul>
</div>
<div class="stage" style="flex-direction: column; justify-content: center; padding: clamp(16px, 2.5vw, 40px);">
<div style="position: relative; width: 100%;">
<button class="copy-btn" onclick="copyTemplate('tpl-framework')" title="Copy template">Copy</button>
<pre id="tpl-framework" class="prompt-block prompt-mid" style="white-space: pre-wrap; font-family: inherit; margin: 0;">Recommended uploads:
1. syllabus.pdf
- Course schedule, objectives, and policies
2. [framework-name].txt
- The analytical method students use
- Write it out in plain language with definitions</pre>
</div>
<div class="tip-box" style="margin-top: 1.2em;">
<p><strong>Consider:</strong> Is there a framework or methodology central to your course? If so, a short document (1-2 pages) explaining it in the terms you use with students could be a strong addition.</p>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 29 — LAYER 2: ASSIGNMENT MATERIALS
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 29: Assignment Materials" data-slide="layer-assignment" tabindex="-1">
<div class="content">
<span class="label label-cyan">Type 2</span>
<h2>Assignment Materials</h2>
<p>These documents define the current task and help the model align its responses with your specific learning objectives.</p>
<ul style="margin-top: 1em; line-height: 1.9;">
<li>What does the assignment ask students to do?</li>
<li>What does strong work on this assignment look like?</li>
<li>What patterns come up most often in your feedback?</li>
</ul>
</div>
<div class="stage" style="flex-direction: column; justify-content: center; padding: clamp(16px, 2.5vw, 40px);">
<div style="position: relative; width: 100%;">
<button class="copy-btn" onclick="copyTemplate('tpl-assignment')" title="Copy template">Copy</button>
<pre id="tpl-assignment" class="prompt-block prompt-mid" style="white-space: pre-wrap; font-family: inherit; margin: 0;">Recommended uploads:
1. [assignment]-prompt.pdf
- The assignment instructions
2. common-feedback.txt
- 5-10 patterns you see every semester
3. strong-examples.txt (optional)
- Excerpts showing what strong work looks like</pre>
</div>
<div class="tip-box" style="margin-top: 1.2em;">
<p><strong>Consider:</strong> Which assignment stands to benefit? Try curating assignment instructions alongside a shortlist of common feedback patterns for starters.</p>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 30 — LAYER 3: SOURCE MATERIALS
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 30: Source Materials" data-slide="layer-sources" tabindex="-1">
<div class="content">
<span class="label label-cyan">Type 3</span>
<h2>Source Materials</h2>
<p>Upload the readings and reference materials students are working with in the current unit. This grounds the model in the actual texts.</p>
<ul style="margin-top: 1em; line-height: 1.9;">
<li>What texts are students reading for this assignment?</li>
<li>Are there reference documents (timelines, glossaries, citation guides)?</li>
<li>Can you add brief metadata or context for each source?</li>
</ul>
</div>
<div class="stage" style="flex-direction: column; justify-content: center; padding: clamp(16px, 2.5vw, 40px);">
<div style="position: relative; width: 100%;">
<button class="copy-btn" onclick="copyTemplate('tpl-sources')" title="Copy template">Copy</button>
<pre id="tpl-sources" class="prompt-block prompt-mid" style="white-space: pre-wrap; font-family: inherit; margin: 0;">Recommended uploads:
1. [reading-title].pdf
- Individual files per text (not one big reader)
- Add a header with: title, author, date, source
2. context-notes.txt (optional)
- 2-3 sentences of context per source
3. [reference-guide].pdf
- Citation style guide, glossary, or timeline</pre>
</div>
<div class="tip-box" style="margin-top: 1.2em;">
<p><strong>Consider:</strong> Which readings or sources are students working with right now? Individual files retrieve better than a single combined PDF.</p>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 31 — CREATE YOUR COLLECTION
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 31: Create Your Collection" data-slide="create-collection" tabindex="-1">
<div class="content">
<span class="label label-cyan">Putting It Together</span>
<h2>Create Your Collection</h2>
<p style="margin-bottom: 0.8em;">Now create the collection you’ve been thinking about.</p>
<div class="card card-cyan step-hidden" data-step="1" style="margin-bottom: 0.5em;">
<h4>1. Go to Workspace → Knowledge</h4>
<p>Click <strong>+ New Knowledge</strong> to start a new collection.</p>
</div>
<div class="card step-hidden" data-step="2" style="margin-bottom: 0.5em;">
<h4>2. Name and describe it</h4>
<p>Give it a clear name tied to your course and a short description of what it contains.</p>
</div>
<div class="card card-navy step-hidden" data-step="3" style="margin-bottom: 0.5em;">
<h4>3. Upload your first file</h4>
<p>Start with the document type that caught your eye - be it course context, assignment materials, or source materials.</p>
</div>
</div>
<div class="stage">
<div class="carousel" data-interval="8000">
<div class="carousel-item active">
<img src="images/slide4-a.png" alt="Workspace Knowledge tab showing collections and New Knowledge button">
<div class="carousel-caption"><strong>Workspace → Knowledge</strong><br>Click <strong>+ New Knowledge</strong> to create a collection.</div>
</div>
<div class="carousel-item">
<img src="images/slide4-b.png" alt="Create a knowledge base form with name, description, and privacy settings">
<div class="carousel-caption"><strong>Name and Describe</strong><br>Give your collection a name and set access to private.</div>
</div>
<div class="carousel-dots"></div>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 32 — ATTACH TO MODEL CARD
========================================== -->
<div class="slide layout-split" role="group" aria-roledescription="slide" aria-label="Slide 32: Attach Your Collection" data-slide="attach-collection" tabindex="-1">
<div class="content">
<span class="label label-cyan">Final Step</span>
<h2>Attach Your Collection</h2>
<p style="margin-bottom: 0.8em;">Connect your collection to the model card you set up earlier.</p>
<div class="card card-cyan step-hidden" data-step="1" style="margin-bottom: 0.5em;">
<h4>1. Return to your model card</h4>
<p>Go to <strong>Workspace → Models</strong> and open the model card you created at the start of today’s session.</p>
</div>
<div class="card step-hidden" data-step="2" style="margin-bottom: 0.5em;">
<h4>2. Attach the collection</h4>
<p>Find the <strong>Knowledge</strong> field and click <strong>Select Knowledge</strong>. Choose the collection you just created.</p>
</div>
<div class="card card-navy step-hidden" data-step="3" style="margin-bottom: 0.5em;">
<h4>3. Save and test</h4>
<p>Save the model card, open a new chat with your model, and ask a question only answerable from your collection.</p>
</div>
</div>
<div class="stage">
<div class="carousel" data-interval="8000">
<div class="carousel-item active">
<img src="images/slide4-c.png" alt="Workspace Models page listing custom models">
<div class="carousel-caption"><strong>Workspace → Models</strong><br>Open the model card you created earlier.</div>
</div>
<div class="carousel-item">
<img src="images/slide4-d.png" alt="Model card editor with Knowledge section highlighted showing Select Knowledge and Upload Files">
<div class="carousel-caption"><strong>Attach Your Collection</strong><br>Click <strong>Select Knowledge</strong> to connect your collection to the model.</div>
</div>
<div class="carousel-dots"></div>
</div>
</div>
</div>
<!-- ==========================================
SLIDE 33 — THE ROAD AHEAD
========================================== -->
<div class="slide layout-full-dark closing-slide" role="group" aria-roledescription="slide" aria-label="Slide 33: The Road Ahead" data-slide="closing" tabindex="-1">