SWE-chat: Coding Agent Interactions From Real Users in the Wild Paper β’ 2604.20779 β’ Published 6 days ago β’ 11 β’ 4
OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation Paper β’ 2604.18486 β’ Published 8 days ago β’ 87 β’ 4
REAM: Merging Improves Pruning of Experts in LLMs Paper β’ 2604.04356 β’ Published 22 days ago β’ 8 β’ 4
Embarrassingly Simple Self-Distillation Improves Code Generation Paper β’ 2604.01193 β’ Published 26 days ago β’ 46 β’ 7
Embarrassingly Simple Self-Distillation Improves Code Generation Paper β’ 2604.01193 β’ Published 26 days ago β’ 46 β’ 7
Omnilingual MT: Machine Translation for 1,600 Languages Paper β’ 2603.16309 β’ Published Mar 17 β’ 21 β’ 5
TERMINATOR: Learning Optimal Exit Points for Early Stopping in Chain-of-Thought Reasoning Paper β’ 2603.12529 β’ Published Mar 13 β’ 19 β’ 3
Accent Vector: Controllable Accent Manipulation for Multilingual TTS Without Accented Data Paper β’ 2603.07534 β’ Published Mar 8 β’ 5 β’ 3
Yor-Sarc: A gold-standard dataset for sarcasm detection in a low-resource African language Paper β’ 2602.18964 β’ Published Feb 21 β’ 1 β’ 4
Late-to-Early Training: LET LLMs Learn Earlier, So Faster and Better Paper β’ 2602.05393 β’ Published Feb 5 β’ 8 β’ 3
Chronicals: A High-Performance Framework for LLM Fine-Tuning with 3.51x Speedup over Unsloth Paper β’ 2601.02609 β’ Published Jan 6 β’ 2 β’ 2
EPAS: Efficient Training with Progressive Activation Sharing Paper β’ 2601.19089 β’ Published Jan 27 β’ 1 β’ 1
Chronicals: A High-Performance Framework for LLM Fine-Tuning with 3.51x Speedup over Unsloth Paper β’ 2601.02609 β’ Published Jan 6 β’ 2 β’ 2