arxiv:2509.20293
Benjamin Feuer PRO
penfever
AI & ML interests
Deep learning, computer vision, large language models, large vision language models
Recent Activity
updated a dataset about 1 hour ago
DCAgent2/eval-DCAgent_a1-bugswarm_DCAgent2_swebench-verified-random-100-folders published a dataset about 1 hour ago
DCAgent2/eval-DCAgent_a1-bugswarm_DCAgent2_swebench-verified-random-100-folders updated a dataset about 2 hours ago
DCAgent2/eval-DCAgent_a1-bash_textbook_DCAgent2_swebench-verified-random-100-foldersOrganizations
models 56
penfever/stageD-thinkbudget-80-8B
8B • Updated • 27
penfever/a3-rl-SankalpKJ_swesmith-oracle-filtered-40-8B
Reinforcement Learning • 8B • Updated • 25
penfever/nl2bash-3k-traces-restore-hp
Text Generation • 308k • Updated • 6
penfever/nl2bash-1k-traces-restore-hp
Text Generation • 308k • Updated • 5
penfever/kimi-k2-swesmith_with_plain_docker-sandboxes-maxeps-32k
Text Generation • 308k • Updated • 5
penfever/GLM-4_6-gemini25flash-stackexchange-overflow-32ep-512k-fixeps
Text Generation • 308k • Updated • 8
penfever/GLM-4_6-taskmaster2-32eps-32k-fixeps
Text Generation • 308k • Updated • 7
penfever/nl2bash-verified-GLM-4_6-traces-32ep-32k-dft
Text Generation • 308k • Updated • 4
penfever/GLM-4_6-inferredbugs-32eps-65k-fixeps
308k • Updated • 3
penfever/glm46-ling-coder-sft-sandboxes-1-maxeps-131k
Text Generation • 308k • Updated • 18 •
datasets 359
penfever/methods2test-large-v2-qwen3.5-122b-32k-traces
Updated
penfever/pymethods2test-large-qwen3.5-122b-32k-traces
Updated • 15 • 1
penfever/nemotron-math-oracle-filtered-qwen3.5-122b-32k-traces
Updated • 19
penfever/mix-h4-binary-easy-qwen3.5-122b-32k-traces
Updated • 11
penfever/delphi-rl-scaling-laws-6279-evals
Updated • 9
penfever/swe_rebench_patched_oracle-minimax-m27-131k-traces_chunk2
Updated • 13
penfever/swe_rebench_patched_oracle-minimax-m27-131k-traces_chunk0
Updated • 13
penfever/swe_rebench_patched_oracle-minimax-m27-131k-traces_chunk1
Updated • 12
penfever/swesmith-oracle-filtered-minimax-m27-131k-traces
Updated • 13
penfever/mix-h2-language-proportional-qwen3.5-122b-32k-traces
Updated • 14