New Links | ML News

ML Newsnew | past | best | rss | submit

1321.	Prism: Demystifying Retention and Interaction in Mid-Training (arxiv.org)
	1 point by xhevahir 197 days ago \| hide \| past \| pdf \| discuss
1322.	Dissociating Direct Access from Inference in AI Introspection (arxiv.org)
	3 points by 3willows 197 days ago \| hide \| past \| pdf \| discuss
1323.	Interwhen: A Generalizable Framework for Verifiable Reasoning (arxiv.org)
	2 points by dimmuborgir 198 days ago \| hide \| past \| pdf \| discuss
1324.	SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels (arxiv.org)
	3 points by matt_d 198 days ago \| hide \| past \| pdf \| discuss
1325.	Transformers Are Bayesian Networks (arxiv.org)
	40 points by Anon84 198 days ago \| hide \| past \| pdf \| 33 comments
1326.	Matrix Valued Residuals (arxiv.org)
	5 points by E-Reverance 198 days ago \| hide \| past \| pdf \| discuss
1327.	MemEvolve: Meta-Evolution of Agent Memory Systems (arxiv.org)
	2 points by dennisy 198 days ago \| hide \| past \| pdf \| 1 comment
1328.	M$^2$RNN: Non-Linear RNNs with Matrix-Valued States for Scalable Modeling (arxiv.org)
	3 points by gmays 198 days ago \| hide \| past \| pdf \| discuss
1329.	Democratizing GraphRAG: Linear, CPU-Only Graph Retrieval for Multi-Hop QA (arxiv.org)
	2 points by PaulHoule 198 days ago \| hide \| past \| pdf \| discuss
1330.	AlgoVeri: An Aligned Benchmark for Verified Code Gen. On Classical Algorithms (arxiv.org)
	2 points by matt_d 198 days ago \| hide \| past \| pdf \| discuss
1331.	The Missing Memory Hierarchy: Demand Paging for LLM Context Windows (arxiv.org)
	3 points by brewcrew 198 days ago \| hide \| past \| pdf \| discuss
1332.	How Do LLMs Compute Verbal Confidence (DeepMind) (arxiv.org)
	3 points by armcat 199 days ago \| hide \| past \| pdf \| discuss
1333.	Evaluating Genuine Reasoning in LLMs via Esoteric Programming Languages (arxiv.org)
	1 point by kerneis 199 days ago \| hide \| past \| pdf \| 1 comment
1334.	Reasoning Core: Procedural Data Generation Suite for Symbolic Pre-Training (arxiv.org)
	2 points by jean-porte 199 days ago \| hide \| past \| pdf \| discuss
1335.	Have LLMs Learned to Reason? A Characterization via 3-SAT Phase Transition (arxiv.org)
	3 points by jacklondon 199 days ago \| hide \| past \| pdf \| 2 comments
1336.	M^2RNN: Non-Linear RNNs with Matrix-Valued States for Scalable Language Modeling (arxiv.org)
	2 points by matt_d 199 days ago \| hide \| past \| pdf \| discuss
1337.	NCCL EP: Towards a Unified Expert Parallel Communication API for NCCL (arxiv.org)
	3 points by matt_d 199 days ago \| hide \| past \| pdf \| discuss
1338.	Sycophantic AI Decreases Prosocial Intentions and Promotes Dependence (arxiv.org)
	3 points by RickJWagner 199 days ago \| hide \| past \| pdf \| discuss
1339.	Do Large Language Models Get Caught in Hofstadter-Mobius Loops? (arxiv.org)
	2 points by pndy 200 days ago \| hide \| past \| pdf \| 1 comment
1340.	Design Conductor: agent autonomously builds a 1.5 GHz Linux-capable RISC-V CPU (arxiv.org)
	4 points by EvgeniyZh 200 days ago \| hide \| past \| pdf \| 1 comment
1341.	SkillNet: Create, Evaluate, and Connect AI Skills (arxiv.org)
	1 point by navikohli 200 days ago \| hide \| past \| pdf \| discuss
1342.	Epiplexity: Rethinking Information for Computationally Bounded Intelligence (arxiv.org)
	2 points by fritzo 200 days ago \| hide \| past \| pdf \| 2 comments
1343.	Mamba-3: Improved Sequence Modeling Using State Space Principles (arxiv.org)
	4 points by anentropic 201 days ago \| hide \| past \| pdf \| 1 comment
1344.	UC Irvine researchers bring down AI powered drones with painted umbrellas (arxiv.org)
	23 points by jcalvinowens 201 days ago \| hide \| past \| pdf \| 8 comments
1345.	Why AI systems don't learn – On autonomous learning from cognitive science (arxiv.org)
	205 points by aanet 201 days ago \| hide \| past \| pdf \| 116 comments
1346.	Pimp My LLM: Leveraging Variability Modeling to Tune Inference Hyperparameters (arxiv.org)
	1 point by PaulHoule 201 days ago \| hide \| past \| pdf \| 1 comment
1347.	Benchmarking Distilled Language Models for Performance and Efficiency (arxiv.org)
	2 points by PaulHoule 201 days ago \| hide \| past \| pdf \| discuss
1348.	Real-World Industrial-Scale Verification: LLM-Driven Theorem Proving on SeL4 (arxiv.org)
	1 point by PaulHoule 201 days ago \| hide \| past \| pdf \| discuss
1349.	We Ran the Largest AI Pokemon Tournament Ever. Now It's an Open Benchmark (arxiv.org)
	1 point by milkkarten 201 days ago \| hide \| past \| pdf \| discuss
1350.	Automating Forecasting Question Generation and Resolution for AI Evaluation (arxiv.org)
	3 points by nbosse 201 days ago \| hide \| past \| pdf \| discuss
	More