Boosting Reasoning via RLVR with Inductive Logic Programming
AI & ML interests
None defined yet.
Recent Activity
View all activity
Do reasoning LLMs actually reason — or learn to game the test? IPT allows for detecting reward hacking in inductive programming tasks (SLR-Bench).
-
Isomorphic Perturbation Testing
🔍1Evaluate rule hypotheses for genuine reasoning vs shortcuts
-
AIML-TUDA/SLR-Bench
Viewer • Updated • 38.5k • 364 • 5 -
SLR-Bench Leaderboard - Reward Hacking in Reasoning Models
🎯1Reward shortcut behavior in LLMs via IPT
-
LLMs Gaming Verifiers: RLVR can Lead to Reward Hacking
Paper • 2604.15149 • Published • 2
-
How to Train your Text-to-Image Model: Evaluating Design Choices for Synthetic Training Captions
Paper • 2506.16679 • Published • 2 -
AIML-TUDA/t2i-diversity-captions
Viewer • Updated • 11M • 144 • 6 -
AIML-TUDA/t2i-diversity-evalprompts
Viewer • Updated • 4.34k • 37 • 3 -
AIML-TUDA/t2i-diversity-gender-neutral-captions
Viewer • Updated • 1M • 18 • 1
A collection of scalable logical reasoning tasks
This collection contains the original repos of the LlavaGuard releases
-
AIML-TUDA/LlavaGuard-v1.2-7B-OV-hf
Image-Text-to-Text • 8B • Updated • 2.43k • 5 -
AIML-TUDA/QwenGuard-v1.2-7B
Image-Text-to-Text • 8B • Updated • 329 • 6 -
AIML-TUDA/LlavaGuard-v1.2-0.5B-OV-hf
Image-Text-to-Text • 0.9B • Updated • 493 • 4 -
AIML-TUDA/QwenGuard-v1.2-3B
Image-Text-to-Text • 4B • Updated • 362 • 4
Boosting Reasoning via RLVR with Inductive Logic Programming
Do reasoning LLMs actually reason — or learn to game the test? IPT allows for detecting reward hacking in inductive programming tasks (SLR-Bench).
-
Isomorphic Perturbation Testing
🔍1Evaluate rule hypotheses for genuine reasoning vs shortcuts
-
AIML-TUDA/SLR-Bench
Viewer • Updated • 38.5k • 364 • 5 -
SLR-Bench Leaderboard - Reward Hacking in Reasoning Models
🎯1Reward shortcut behavior in LLMs via IPT
-
LLMs Gaming Verifiers: RLVR can Lead to Reward Hacking
Paper • 2604.15149 • Published • 2
A collection of scalable logical reasoning tasks
-
How to Train your Text-to-Image Model: Evaluating Design Choices for Synthetic Training Captions
Paper • 2506.16679 • Published • 2 -
AIML-TUDA/t2i-diversity-captions
Viewer • Updated • 11M • 144 • 6 -
AIML-TUDA/t2i-diversity-evalprompts
Viewer • Updated • 4.34k • 37 • 3 -
AIML-TUDA/t2i-diversity-gender-neutral-captions
Viewer • Updated • 1M • 18 • 1
This collection contains the original repos of the LlavaGuard releases
-
AIML-TUDA/LlavaGuard-v1.2-7B-OV-hf
Image-Text-to-Text • 8B • Updated • 2.43k • 5 -
AIML-TUDA/QwenGuard-v1.2-7B
Image-Text-to-Text • 8B • Updated • 329 • 6 -
AIML-TUDA/LlavaGuard-v1.2-0.5B-OV-hf
Image-Text-to-Text • 0.9B • Updated • 493 • 4 -
AIML-TUDA/QwenGuard-v1.2-3B
Image-Text-to-Text • 4B • Updated • 362 • 4