conference-paper-writing

Name: conference-paper-writing
Author: Boom5426

51installs28forks

How to install conference-paper-writing?

boom5426/nature-paper-skills/conference-paper-writing

$npx -y skills add Boom5426/Nature-Paper-Skills --skill conference-paper-writing

Installs into the current project.

Summary

Use when writing or revising ML or AI conference papers for venues such as NeurIPS, ICML, ICLR, ACL, AAAI, or COLM, especially when the workflow is conference-first rather than Nature-style journal-first.

Files · 1

View on GitHub

SKILL.md

1	# Conference Paper Writing for Top AI Venues
2
3	Expert-level guidance for writing publication-ready papers targeting NeurIPS, ICML, ICLR, ACL, AAAI, and COLM. This skill combines writing philosophy from top researchers (Nanda, Farquhar, Karpathy, Lipton, Steinhardt) with practical tools: LaTeX templates, citation verification APIs, and conference checklists.
4
5	## Core Philosophy: Collaborative Writing
6
7	Paper writing is collaborative, but Claude should be proactive in delivering drafts.
8
9	The typical workflow starts with a research repository containing code, results, and experimental artifacts. Claude's role is to:
10
11	1. Understand the project by exploring the repo, results, and existing documentation
12	2. Deliver a complete first draft when confident about the contribution
13	3. Search literature using web search and APIs to find relevant citations
14	4. Refine through feedback cycles when the scientist provides input
15	5. Ask for clarification only when genuinely uncertain about key decisions
16
17	Key Principle: Be proactive. If the repo and results are clear, deliver a full draft. Don't block waiting for feedback on every section—scientists are busy. Produce something concrete they can react to, then iterate based on their response.
18
19	---
20
21	## ⚠️ CRITICAL: Never Hallucinate Citations
22
23	This is the most important rule in academic writing with AI assistance.
24
25	### The Problem
26	AI-generated citations have a ~40% error rate. Hallucinated references—papers that don't exist, wrong authors, incorrect years, fabricated DOIs—are a serious form of academic misconduct that can result in desk rejection or retraction.
27
28	### The Rule
29	NEVER generate BibTeX entries from memory. ALWAYS fetch programmatically.
30
31	\| Action \| ✅ Correct \| ❌ Wrong \|
32	\|--------\|-----------\|----------\|
33	\| Adding a citation \| Search API → verify → fetch BibTeX \| Write BibTeX from memory \|
34	\| Uncertain about a paper \| Mark as `[CITATION NEEDED]` \| Guess the reference \|
35	\| Can't find exact paper \| Note: "placeholder - verify" \| Invent similar-sounding paper \|
36
37	### When You Can't Verify a Citation
38
39	If you cannot programmatically verify a citation, you MUST:
40
41	```latex
42	% EXPLICIT PLACEHOLDER - requires human verification
43	\cite{PLACEHOLDER_author2024_verify_this} % TODO: Verify this citation exists
44	```
45
46	Always tell the scientist: "I've marked [X] citations as placeholders that need verification. I could not confirm these papers exist."
47
48	### Recommended: Install Exa MCP for Paper Search
49
50	For the best paper search experience, install Exa MCP which provides real-time academic search:
51
52	Claude Code:
53	```bash
54	claude mcp add exa -- npx -y mcp-remote "https://mcp.exa.ai/mcp"
55	```
56
57	Cursor / VS Code (add to MCP settings):
58	```json
59	{
60	"mcpServers": {
61	"exa": {
62	"type": "http",
63	"url": "https://mcp.exa.ai/mcp"
64	}
65	}
66	}
67	```
68
69	Exa MCP enables searches like:
70	- "Find papers on RLHF for language models published after 2023"
71	- "Search for transformer architecture papers by Vaswani"
72	- "Get recent work on sparse autoencoders for interpretability"
73
74	Then verify results with Semantic Scholar API and fetch BibTeX via DOI.
75
76	---
77
78	## Workflow 0: Starting from a Research Repository
79
80	When beginning paper writing, start by understanding the project:
81
82	```
83	Project Understanding:
84	- [ ] Step 1: Explore the repository structure
85	- [ ] Step 2: Read README, existing docs, and key results
86	- [ ] Step 3: Identify the main contribution with the scientist
87	- [ ] Step 4: Find papers already cited in the codebase
88	- [ ] Step 5: Search for additional relevant literature
89	- [ ] Step 6: Outline the paper structure together
90	- [ ] Step 7: Draft sections iteratively with feedback
91	```
92
93	Step 1: Explore the Repository
94
95	```bash
96	# Understand project structure
97	ls -la
98	find . -name "*.py" \| head -20
99	find . -name ".md" -o -name ".txt" \| xargs grep -l -i "result\\|conclusion\\|finding"
100	```
101
102	Look for:
103	- `README.md` - Project overview and claims
104	- `results/`, `outputs/`, `experiments/` - Key findings
105	- `configs/` - Experimental settings
106	- Existing `.bib` files or citation references
107	- Any draft documents or notes
108
109	Step 2: Identify Existing Citations
110
111	Check for papers already referenced in the codebase:
112
113	```bash
114	# Find existing citations
115	grep -r "arxiv\\|doi\\|cite" --include=".md" --include=".bib" --include="*.py"
116	find . -name "*.bib"
117	```
118
119	These are high-signal starting points for Related Work—the scientist has already deemed them relevant.
120
121	Step 3: Clarify the Contribution
122
123	Before writing, explicitly confirm with the scientist:
124
125	> "Based on my understanding of the repo, the main contribution appears to be [X].
126	> The key results show [Y]. Is this the framing you want for the paper,
127	> or should we emphasize dif

People who liked this also viewed

6 picks

Type

conference-paper-writing

51installs28forks

View on GitHub

How to install conference-paper-writing?

boom5426/nature-paper-skills/conference-paper-writing

$npx -y skills add Boom5426/Nature-Paper-Skills --skill conference-paper-writing

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Conference Paper Writing for Top AI Venues
2
3	Expert-level guidance for writing publication-ready papers targeting NeurIPS, ICML, ICLR, ACL, AAAI, and COLM. This skill combines writing philosophy from top researchers (Nanda, Farquhar, Karpathy, Lipton, Steinhardt) with practical tools: LaTeX templates, citation verification APIs, and conference checklists.
4
5	## Core Philosophy: Collaborative Writing
6
7	Paper writing is collaborative, but Claude should be proactive in delivering drafts.
8
9	The typical workflow starts with a research repository containing code, results, and experimental artifacts. Claude's role is to:
10
11	1. Understand the project by exploring the repo, results, and existing documentation
12	2. Deliver a complete first draft when confident about the contribution
13	3. Search literature using web search and APIs to find relevant citations
14	4. Refine through feedback cycles when the scientist provides input
15	5. Ask for clarification only when genuinely uncertain about key decisions
16
17	Key Principle: Be proactive. If the repo and results are clear, deliver a full draft. Don't block waiting for feedback on every section—scientists are busy. Produce something concrete they can react to, then iterate based on their response.
18
19	---
20
21	## ⚠️ CRITICAL: Never Hallucinate Citations
22
23	This is the most important rule in academic writing with AI assistance.
24
25	### The Problem
26	AI-generated citations have a ~40% error rate. Hallucinated references—papers that don't exist, wrong authors, incorrect years, fabricated DOIs—are a serious form of academic misconduct that can result in desk rejection or retraction.
27
28	### The Rule
29	NEVER generate BibTeX entries from memory. ALWAYS fetch programmatically.
30
31	\| Action \| ✅ Correct \| ❌ Wrong \|
32	\|--------\|-----------\|----------\|
33	\| Adding a citation \| Search API → verify → fetch BibTeX \| Write BibTeX from memory \|
34	\| Uncertain about a paper \| Mark as `[CITATION NEEDED]` \| Guess the reference \|
35	\| Can't find exact paper \| Note: "placeholder - verify" \| Invent similar-sounding paper \|
36
37	### When You Can't Verify a Citation
38
39	If you cannot programmatically verify a citation, you MUST:
40
41	```latex
42	% EXPLICIT PLACEHOLDER - requires human verification
43	\cite{PLACEHOLDER_author2024_verify_this} % TODO: Verify this citation exists
44	```
45
46	Always tell the scientist: "I've marked [X] citations as placeholders that need verification. I could not confirm these papers exist."
47
48	### Recommended: Install Exa MCP for Paper Search
49
50	For the best paper search experience, install Exa MCP which provides real-time academic search:
51
52	Claude Code:
53	```bash
54	claude mcp add exa -- npx -y mcp-remote "https://mcp.exa.ai/mcp"
55	```
56
57	Cursor / VS Code (add to MCP settings):
58	```json
59	{
60	"mcpServers": {
61	"exa": {
62	"type": "http",
63	"url": "https://mcp.exa.ai/mcp"
64	}
65	}
66	}
67	```
68
69	Exa MCP enables searches like:
70	- "Find papers on RLHF for language models published after 2023"
71	- "Search for transformer architecture papers by Vaswani"
72	- "Get recent work on sparse autoencoders for interpretability"
73
74	Then verify results with Semantic Scholar API and fetch BibTeX via DOI.
75
76	---
77
78	## Workflow 0: Starting from a Research Repository
79
80	When beginning paper writing, start by understanding the project:
81
82	```
83	Project Understanding:
84	- [ ] Step 1: Explore the repository structure
85	- [ ] Step 2: Read README, existing docs, and key results
86	- [ ] Step 3: Identify the main contribution with the scientist
87	- [ ] Step 4: Find papers already cited in the codebase
88	- [ ] Step 5: Search for additional relevant literature
89	- [ ] Step 6: Outline the paper structure together
90	- [ ] Step 7: Draft sections iteratively with feedback
91	```
92
93	Step 1: Explore the Repository
94
95	```bash
96	# Understand project structure
97	ls -la
98	find . -name "*.py" \| head -20
99	find . -name ".md" -o -name ".txt" \| xargs grep -l -i "result\\|conclusion\\|finding"
100	```
101
102	Look for:
103	- `README.md` - Project overview and claims
104	- `results/`, `outputs/`, `experiments/` - Key findings
105	- `configs/` - Experimental settings
106	- Existing `.bib` files or citation references
107	- Any draft documents or notes
108
109	Step 2: Identify Existing Citations
110
111	Check for papers already referenced in the codebase:
112
113	```bash
114	# Find existing citations
115	grep -r "arxiv\\|doi\\|cite" --include=".md" --include=".bib" --include="*.py"
116	find . -name "*.bib"
117	```
118
119	These are high-signal starting points for Related Work—the scientist has already deemed them relevant.
120
121	Step 3: Clarify the Contribution
122
123	Before writing, explicitly confirm with the scientist:
124
125	> "Based on my understanding of the repo, the main contribution appears to be [X].
126	> The key results show [Y]. Is this the framing you want for the paper,
127	> or should we emphasize dif

People who liked this also viewed

6 picks

Type