September 13, 2025 front

ML Newsnew | past | best | rss | submit

Stories from September 13, 2025 (UTC)

Go back a day, month, or year. Go forward a day.

1.	Instruction-Following Pruning for Large Language Models (arxiv.org)
	5 points by gok on Sep 13, 2025 \| hide \| past \| pdf \| discuss
2.	Reverse-Engineered Reasoning for Open-Ended Generation (arxiv.org)
	4 points by amrrs on Sep 13, 2025 \| hide \| past \| pdf \| 1 comment
3.	The SWE-Bench Illusion: When LLMs Remember Instead of Reason (arxiv.org)
	4 points by nurple on Sep 13, 2025 \| hide \| past \| pdf \| discuss
4.	Lossless Compression of Neural Network Components in Low-Precision Formats (arxiv.org)
	3 points by PaulHoule on Sep 13, 2025 \| hide \| past \| pdf \| discuss
5.	Correctness-Guaranteed Code Generation via Constrained Decoding (arxiv.org)
	3 points by PaulHoule on Sep 13, 2025 \| hide \| past \| pdf \| discuss
6.	Optimization Pathways for Long-Context Agentic LLM Inference (arxiv.org)
	3 points by ceolin on Sep 13, 2025 \| hide \| past \| pdf \| discuss
7.	RL's Razor: Why Online Reinforcement Learning Forgets Less (arxiv.org)
	3 points by Anon84 on Sep 13, 2025 \| hide \| past \| pdf \| discuss
8.	The Illusion of Diminishing Returns: Measuring Long Horizon Execution in LLMs (arxiv.org)
	3 points by shash42 on Sep 13, 2025 \| hide \| past \| pdf \| 1 comment