August 14, 2026 front

ML Newsnew | past | best | rss | submit

Stories from August 14, 2026 (UTC)

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

1.	A Contract-Grade Verifier for LLM-Generated GPU Kernels (arxiv.org)
	46 points by Jimmc414 51 days ago \| hide \| past \| pdf \| discuss
2.	CAKE: Compiler-Agent Co-Design for Frontier Kernel Evolution (arxiv.org)
	8 points by matt_d 51 days ago \| hide \| past \| pdf \| discuss
3.	Towards a Risk Assessment of Malicious Skill Files in Coding Agents (arxiv.org)
	5 points by Beko2210 51 days ago \| hide \| past \| pdf \| discuss
4.	V-RAE: Rethinking Video Latent Spaces for Generation (arxiv.org)
	3 points by selimonder 51 days ago \| hide \| past \| pdf \| discuss
5.	Hypothetical Document Embeddings (arxiv.org)
	3 points by kaycebasques 51 days ago \| hide \| past \| pdf \| discuss
6.	Adversarial Paraphrasing: Attack for Humanizing AI-Generated Text (2025) (arxiv.org)
	2 points by 0x_rs 51 days ago \| hide \| past \| pdf \| discuss
7.	Reasoning Jury: Multi-Model Consensus for Evaluating Reasoning Traces (arxiv.org)
	2 points by sonabinu 51 days ago \| hide \| past \| pdf \| discuss
8.	We tested every venue in 2 Bali regions against 4 AI assistants (arxiv.org)
	2 points by fspunch 51 days ago \| hide \| past \| pdf \| discuss
9.	StoryScope: Investigating Idiosyncrasies in AI Fiction (arxiv.org)
	1 point by dgellow 51 days ago \| hide \| past \| pdf \| discuss
10.	A Probe Direction Is a Property of Its Prompt (arxiv.org)
	1 point by sbulaev 51 days ago \| hide \| past \| pdf \| discuss
11.	Agent Safety Should Be a Runtime Contract (arxiv.org)
	1 point by sbulaev 51 days ago \| hide \| past \| pdf \| discuss
12.	Why Does Claude.md Keep Growing? Catastrophic Remembering in Agentic Coding (arxiv.org)
	1 point by oskrim 51 days ago \| hide \| past \| pdf \| 1 comment