When Models Edit Too Much: On the Fidelity of Minimal Code Edits
Paper • 2609.04061 • Published • 5
M Saad Salman
MSS444
AI & ML interests
None yet
Recent Activity
upvoted a paper 38 minutes ago
When Models Edit Too Much: On the Fidelity of Minimal Code Edits upvoted a paper 39 minutes ago
Iris: Climbing to the Search Frontier upvoted a paper 39 minutes ago
Bilevel Coordinated Reflection: A Game-Theoretic Approach to Multi-Agent LLM SystemsOrganizations
None yet
upvoted a paper 38 minutes ago
upvoted 2
papers 39 minutes ago
upvoted 2
papers 4 days ago
upvoted 7
papers 5 days ago
PaperCompiler: Faithful Paper-to-Code Generation via Repository-Level Specification Compilation
Paper • 2609.02272 • Published • 6
S3Gym: Can LLMs Turn Self-Testing and Self-Judging into Self-Improvement?
Paper • 2608.31100 • Published • 38
Language Models Can Control Their Own Attention
Paper • 2609.02737 • Published • 65
HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?
Paper • 2609.01437 • Published • 256
Aspire: Can Models Self-Evolve from Vague Goals?
Paper • 2608.31111 • Published • 222
EarlyEval: Cheaper Agent Evaluation via Early Outcome Prediction
Paper • 2609.02783 • Published • 115
Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills
Paper • 2609.02749 • Published • 529
upvoted 3
papers 6 days ago
upvoted 3
papers 7 days ago
CAST: Critique-Aware Supervision for Training Reliable Long-Horizon Tool-Calling Agents
Paper • 2608.30147 • Published • 8
Learning to Evaluate Before Improving: Automatic Rubric Induction for Automatic Research Agents
Paper • 2608.31076 • Published • 20
Scaling Large Reasoning Models beyond Human Supervision: A Path toward Superintelligence
Paper • 2608.31075 • Published • 28
upvoted 2
papers 8 days ago