Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?
Paper • 2603.24472 • Published • 48
Byung-Kwan Lee
BK-Lee
AI & ML interests
Vision-Language Models
Recent Activity
upvoted a paper 9 days ago
Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs? upvoted a paper 15 days ago
Efficient Reasoning with Balanced ThinkingOrganizations
upvoted a paper 9 days ago
upvoted 3
papers 15 days ago
Efficient Reasoning with Balanced Thinking
Paper • 2603.12372 • Published • 143
Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation
Paper • 2603.19220 • Published • 65
ProRL Agent: Rollout-as-a-Service for RL Training of Multi-Turn LLM Agents
Paper • 2603.18815 • Published • 14
upvoted a paper about 1 month ago
upvoted 9
papers 3 months ago
OpenVoxel: Training-Free Grouping and Captioning Voxels for Open-Vocabulary 3D Scene Understanding
Paper • 2601.09575 • Published • 26
Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning
Paper • 2601.09708 • Published • 54
GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
Paper • 2601.05242 • Published • 229
SurgWorld: Learning Surgical Robot Policies from Videos via World Modeling
Paper • 2512.23162 • Published • 14
Quantile Rendering: Efficiently Embedding High-dimensional Feature on 3D Gaussian Splatting
Paper • 2512.20927 • Published • 17
Masking Teacher and Reinforcing Student for Distilling Vision-Language Models
Paper • 2512.22238 • Published • 30
NVIDIA Nemotron 3: Efficient and Open Intelligence
Paper • 2512.20856 • Published • 43
Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
Paper • 2512.20848 • Published • 42
JustRL: Scaling a 1.5B LLM with a Simple RL Recipe
Paper • 2512.16649 • Published • 27
upvoted 6
papers 4 months ago
Zoom-Zero: Reinforced Coarse-to-Fine Video Understanding via Temporal Zoom-in
Paper • 2512.14273 • Published • 10
DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
Paper • 2512.02556 • Published • 265
Qwen3-VL Technical Report
Paper • 2511.21631 • Published • 161
RefineBench: Evaluating Refinement Capability of Language Models via Checklists
Paper • 2511.22173 • Published • 15
OpenMMReasoner: Pushing the Frontiers for Multimodal Reasoning with an Open and General Recipe
Paper • 2511.16334 • Published • 95
Nemotron Elastic: Towards Efficient Many-in-One Reasoning LLMs
Paper • 2511.16664 • Published • 29