New Links | ML News

ML Newsnew | past | best | rss | submit

1561.	Routing LLM queries using internal success predictions (70% cost reduction) (arxiv.org)
	1 point by stansApprentice 234 days ago \| hide \| past \| pdf \| 3 comments
1562.	SWE-AGI: benchmarking spec-driven software construction (arxiv.org)
	1 point by mustaphah 234 days ago \| hide \| past \| pdf \| 1 comment
1563.	RL on GPT-5 to write better kernels (arxiv.org)
	4 points by atallahw 235 days ago \| hide \| past \| pdf \| 1 comment
1564.	Pushing Tensor Accelerators Beyond MatMul in a User-Schedulable Language (arxiv.org)
	2 points by matt_d 235 days ago \| hide \| past \| pdf \| discuss
1565.	HySparse: A Hybrid Sparse Attention Architecture (arxiv.org)
	5 points by readitalready 235 days ago \| hide \| past \| pdf \| discuss
1566.	Biases in the Blind Spot: Detecting What LLMs Fail to Mention (arxiv.org)
	1 point by jari_mustonen 235 days ago \| hide \| past \| pdf \| discuss
1567.	Evaluation of RAG Architectures for Policy Document Question Answering (arxiv.org)
	1 point by PaulHoule 235 days ago \| hide \| past \| pdf \| discuss
1568.	SoftMatcha 2: A Fast and Soft Pattern Matcher for Trillion-Scale Corpora (arxiv.org)
	3 points by salkahfi 235 days ago \| hide \| past \| pdf \| discuss
1569.	Opus: Towards Efficient and Principled Data Selection in LLM Pre-Training (arxiv.org)
	2 points by onurkanbkrc 235 days ago \| hide \| past \| pdf \| discuss
1570.	Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters (arxiv.org)
	1 point by onurkanbkrc 235 days ago \| hide \| past \| pdf \| 1 comment
1571.	Faster and Cheaper Computations with Randomized Numerical Linear Algebra (arxiv.org)
	2 points by PaulHoule 235 days ago \| hide \| past \| pdf \| discuss
1572.	NanoQuant: Efficient Sub-1-Bit Quantization of Large Language Models (arxiv.org)
	13 points by chrsw 235 days ago \| hide \| past \| pdf \| discuss
1573.	Grok4 sabotages shutdown 97% of the time,even if instructed not in system prompt (arxiv.org)
	8 points by agenticagent 235 days ago \| hide \| past \| pdf \| 4 comments
1574.	Biases in the Blind Spot: Detecting What LLMs Fail to Mention (arxiv.org)
	4 points by typeofhuman 235 days ago \| hide \| past \| pdf \| 1 comment
1575.	Attention Sinks and Compression Valleys in LLMs (arxiv.org)
	1 point by alexkranias 236 days ago \| hide \| past \| pdf \| discuss
1576.	Misconduct in Post-Selections and Deep Learning (2024) (arxiv.org)
	3 points by bjourne 236 days ago \| hide \| past \| pdf \| discuss
1577.	The Hot Mess of AI: How Does Misalignment Scale with Model Intelligence (arxiv.org)
	1 point by schmuhblaster 236 days ago \| hide \| past \| pdf \| discuss
1578.	GRP-Obliteration: Unaligning LLMs with a Single Unlabeled Prompt [pdf] (arxiv.org)
	2 points by janandonly 236 days ago \| hide \| past \| pdf \| discuss
1579.	Harmless reward hacks generalize to shutdown evasion and dictatorship in GPT-4.1 (arxiv.org)
	1 point by toliveistobuild 236 days ago \| hide \| past \| pdf \| 1 comment
1580.	FullStack-Agent: Enhancing Agentic Full-Stack Web Coding (arxiv.org)
	2 points by simonpure 237 days ago \| hide \| past \| pdf \| discuss
1581.	Lightweight Memory Construction with Dynamic Evolution for LLM Agents (arxiv.org)
	2 points by PaulHoule 237 days ago \| hide \| past \| pdf \| discuss
1582.	Moltbook: Fast Response or Silence? (arxiv.org)
	1 point by EagleEdge 237 days ago \| hide \| past \| pdf \| discuss
1583.	Randomness in Agentic Evals (arxiv.org)
	3 points by andre15silva 237 days ago \| hide \| past \| pdf \| discuss
1584.	Large Language Model Reasoning Failures (arxiv.org)
	1 point by mpweiher 237 days ago \| hide \| past \| pdf \| discuss
1585.	We Should Separate Memorization from Copyright (arxiv.org)
	1 point by 50kIters 237 days ago \| hide \| past \| pdf \| discuss
1586.	Security audit of Browser Use: prompt injection, credential exfil, domain bypass (arxiv.org)
	2 points by tiny-automates 237 days ago \| hide \| past \| pdf \| 1 comment
1587.	Frontier AI agents violate ethical constraints 30–50% of time, pressured by KPIs (arxiv.org)
	544 points by tiny-automates 237 days ago \| hide \| past \| pdf \| 366 comments
1588.	Large Language Model Reasoning Failures (arxiv.org)
	3 points by belter 238 days ago \| hide \| past \| pdf \| discuss
1589.	Shared LoRA Subspaces for Almost Strict Continual Learning (arxiv.org)
	1 point by unisub_guy 238 days ago \| hide \| past \| pdf \| 1 comment
1590.	Towards Understanding What State Space Models Learn About Code (arxiv.org)
	1 point by belter 238 days ago \| hide \| past \| pdf \| discuss
	More