Hackernews posts about arXiv.org

ArXiv.ORG is an online repository that provides open access to electronic preprints of completed peer-reviewed papers in physics, mathematics, computer science, and related disciplines.

RSS feed for arXiv.org

Related:HTML5 Einstein

121
newHybrid Reasoning Agents: LLM agents that think in latent space, act through textarxiv.org
▲ 520allinonai 20d 💬 239
122
newSleep framework: A new way for LLMs to retain knowledge in continual learningresearch.google
▲ 277acossta 10d 💬 58
123
newShow HN: T0: We trained a 256M SOTA open-weights time-series forecasting modelgithub.com
▲ 14LTMeyer 24d discuss
124
newAsk HN: Is anyone running multi/always-on agents that don't cost a fortune?
▲ 6abejora 46h 💬 5
125
newShow HN: Rufous – Agent attuned to small modelsrufous.chat
▲ 3stephanfebs 6d discuss
126
newShow HN: simdown – use SIMD to parse Markdown at ~1 GB/sgithub.com
▲ 2trevorah 13d discuss
127
newAsk HN: Are Models Getting Better at Software Estimations of Time/Effort/Cost?
▲ 2artur_makly 17d discuss
128
newShow HN: An improved kokoro distillation in 9MBhuggingface.co
▲ 1mithrilman 23d discuss
129
newThe Little Book of Generative AI Foundations: An Intuitive Mathematical Primerarxiv.org
▲ 393ibobev 4d 💬 60
130
newAI models can provably hide arbitrary capabilitiesarxiv.org
▲ 282sbulaev 26d 💬 145
131
newWhy is AI generated language so hard to understand?arxiv.org
▲ 259canjobear 14d 💬 205
132
newLeaking Secrets from Air-Gap Computers by Spelling Radio Signals from RAM (2024)arxiv.org
▲ 222hapax_legomena 7d 💬 106
133
newThe alien space of science: Coherent but cognitively unavailable directionsarxiv.org
▲ 214martincsweiss 16d 💬 89
134
newAttention Is Not What You Need (2025)arxiv.org
▲ 197westurner 13d 💬 72
135
newProving and improving the Platonic Representation Hypothesis with JEPAsarxiv.org
▲ 172Anon84 6d 💬 85
136
newAI and the Enclosure of the Scientific Commonsarxiv.org
▲ 142jrepinc 16d 💬 81
137
newMemorilla: Latent Semantic Memory for LLMsarxiv.org
▲ 116handfuloflight 40h 💬 13
138
newAgentic backdoors via pretraining poisoningarxiv.org
▲ 108gmays 5d 💬 31
139
newOpen-Source 5G Ran Platforms: A Dual Perspective on Performance and Capabilitiesarxiv.org
▲ 54teleforce 5d 💬 5
140
newUSDB: Scaling Storage for Next-Generation Recommender Systems at ByteDancearxiv.org
▲ 491equalsequals1 4d 💬 8
141
newExecutable Archaeology: Reanimating the Heuristic Compiler (1961) and KBEmacs (1984)arxiv.org
▲ 34abrax3141 11d discuss
142
newWhen Modernization Breaks Working Code: A Study of LLMs on Legacy PHParxiv.org
▲ 22tambourine_man 6d 💬 14
143
newMagic state cultivation on a superconducting quantum processorarxiv.org
▲ 18kheal 12d 💬 1
144
newSelf-Evolving Agent Harnessarxiv.org
▲ 18axis_official 23d 💬 1
145
newMonoidal computer I: Basic computability by string diagrams (2012)arxiv.org
▲ 16measurablefunc 16d discuss
146
newThe Illusion of Thinking: Models Don't Actually Reasonarxiv.org
▲ 14CGMthrowaway 9d 💬 1
147
newCompaction Laundering: Provenance Destruction in Context-Compacting Coding Agentarxiv.org
▲ 9nahsra 12d 💬 2
148
newThe Thinking Leakage in Reinforcement Learning for LLMsarxiv.org
▲ 8ilreb 23d discuss
149
newA Strongly Aperiodic Monotile in Three Dimensionsarxiv.org
▲ 7nhatcher 26d 💬 7
150
newMulti-Token Prediction Done Rightarxiv.org
▲ 7matt_d 4d 💬 2

More stories →