11-785 Project Grading Information

	A	B	C	D	E	F	G
1	Here is the template you are required to use for report submission: https://www.overleaf.com/read/btddtdztdrgs		Proposal (1–2 pages)	Midterm Report (4–5 pages)	Final Report (8–10 pages)		↴

2	Overview and Context	Abstract	The problem statement is clear and concise, understandable to a general audience.	The problem statement is understandable to non-experts.	The problem statement is clearly articulated and understandable to a non-expert audience.		Regarding the first submission -- the Project Proposal Keep your proposal concise:
3	Overview and Context	Motivation & Objectives	Students briefly explain why the problem is important or relevant and outline at least one broad goal or objective (even if quantitative details are not yet finalized).	Students articulate why the problem is important or who benefits and state at least one qualitative or quantitative objective.	Contextual motivation is provided and qualitative and quantitative objectives are clearly stated and related to the problem.		- You are trying to pitch an idea and you should convey your thoughts as comprehensively as possible but while being concise. - Try to limit to 2 pages unless you have done some significant work, or have an extensive literature review (which we do highly support that you start working on). - The more papers you read the better chances you have to come up with novel ideas that can be published.
4	Related Work and Background	Literature Review	Students cite at least one prior work relevant to their project. A short explanation of how this work relates to their project is included.	Students cite 1–2 key papers related to their project, with brief connections to the problem.	Students identify relevant prior work, preferably grouping related research lines or methodologies. Proper citation and clear connections to the problem are included.
5	Related Work and Background	Background	If applicable, students may reference an evaluation metric or baseline related to their problem.	At least one cited paper connects the evaluation metric or baseline to the problem.	Cited papers explicitly connect the evaluation metric and baseline models to the problem statement.
6	Methodology	Model Description	A high-level description of the approach or model is provided. Diagrams/tables optional but encouraged.	Basic explanation of the proposed model (inputs/outputs). Diagrams/tables encouraged.	Clear diagrams/tables describing inputs, outputs, and structure. Output map shapes and parameter counts included.	→	Example of Model Description Table From the following article: https://www.researchgate.net/publication/333248441_Accurate_prediction_of_boundaries_of_high_resolution_topologically_associated_domains_TADs_in_fruit_flies_using_deep_learning
7		Dataset	Students identify a potential data source and briefly describe how it relates to the problem.	Data source is identified and basic preparation steps outlined.	Steps for training and evaluation are explicitly stated, including batch sampling. Dataset citation provided.
8		Evaluation Metric	The metric students plan to use for assessing success is mentioned.	The chosen metric is stated, with a simple explanation of its relevance.	Mathematical definitions with variables clearly labeled and tied to the problem.
9		Loss Function	Not required but may be mentioned.	Mentioned briefly if applicable.	Loss and optimization described with mathematical precision.
10		Experimental Depth & Iteration	Students state at least one planned model/experiment variation beyond the baseline.	Students present evidence of initial iterations (e.g., at least two runs, adjustments, or configurations attempted).	Students demonstrate multiple experimental attempts or configurations, showing development beyond a single training run.
11	Baseline and Extensions	Baseline Selection & Evaluation	Students propose one or more potential baselines for comparison.	MANDATORY: Students implement and explain their chosen baseline.	Students justify their choice of baseline, referencing relevant papers or models. Baseline evaluations are conducted systematically.
12		Implemented Extensions/ Experiments	Students describe 1–2 initial ideas for how their work might extend beyond the baseline.	High-level description of extensions/experiments to be conducted.	Results show improvement over baselines or justified analysis of failures.
13		Baseline Reproduction Evidence		Students show initial results/plots confirming that their baseline results are their own.	Students provide clear, original baseline output (plots/logs/metrics) confirming reproduction.
14	Results and Analysis	Results	Anticipated results described in general terms (e.g., expected improvement over baseline). Not required to include validation results.	Preliminary results or progress metrics encouraged (e.g., sample outputs or early plots).	Separate and clear results for training and validation (plots/tables). Key findings visualized.
15		Error / Failure Case Analysis			At least one failure mode or error pattern is identified and discussed.
16		Sensitivity / Ablation Analysis			At least one comparison demonstrating how performance changes with a design choice (e.g., model variant, hyperparameter).
17	Discussion		Students briefly discuss potential challenges or risks in executing the proposed project.	Students briefly discuss challenges so far and potential risks moving forward.	The team explains the significance of results, potential risks, and sensitivity of results to input changes, discussing limitations where applicable.
18	Future Directions		Planned Work: A clear outline of next steps to be done before the midterm report.	Planned Next Steps: Students identify 2–3 specific actions to move forward.	Students identify areas for future exploration or improvement based on results and limitations.
19	Conclusion				Students effectively summarize their key findings, progress made, and how the work relates to their objectives.
20	Bonus Contributions (Optional)	Visualization			Compelling and clearly labeled visualizations (embeddings, features) may earn extra points.
21	Bonus Contributions (Optional)	Extra Experimental Exploration			Significant additional experiments or analyses beyond required work may earn extra points.
22	Administrative Details	Bibliography	At least one citation (preferably two).	Students include citations for all referenced work (2–3 minimum).	References formatted correctly and thoughtfully selected.
23		Team Contributions	If applicable, list initial roles/responsibilities.	Provide basic role breakdown.	Provide contribution breakdown by member.
24		GitHub	Begin a GitHub repository if modeling has started and include link.	Include GitHub link.	Include GitHub link.	→	In the case that one+ member of your team does not contribute enough to the team project, we will use the GitHub history (as well as google docs and Overleaf) to assess each team member's contribution level.
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