Show HN: Mars โ€“ Personal AI robot for builders (< $2k)
news.ycombinator.comยท4hยท
Discuss: Hacker News
๐ŸŽฎSwitch Archaeology
We built AI to split the monolith
engineering.monday.comยท10hยท
Discuss: Hacker News
๐Ÿ”„Language Evolution
Vibe engineering
simonwillison.netยท4hยท
๐Ÿ“Code Metrics
Token Hidden Reward: Steering Exploration-Exploitation in Group Relative Deep Reinforcement Learning
arxiv.orgยท15h
๐Ÿ”ฒCellular Automata
LLM Data Detox: Erasing the Past for a Brighter AI Future
dev.toยท5hยท
Discuss: DEV
๐Ÿ’ปLocal LLMs
DiffuSpec: Unlocking Diffusion Language Models for Speculative Decoding
arxiv.orgยท1d
๐ŸŽ™๏ธWhisper
Beyond Outcome Reward: Decoupling Search and Answering Improves LLM Agents
arxiv.orgยท15h
โšกProof Automation
Psychological Steering in LLMs: An Evaluation of Effectiveness and Trustworthiness
arxiv.orgยท15h
๐ŸŽฏThreat Hunting
TS-Reasoner: Aligning Time Series Foundation Models with LLM Reasoning
arxiv.orgยท15h
๐ŸŽฏGradual Typing
StaMo: Unsupervised Learning of Generalizable Robot Motion from Compact State Representation
arxiv.orgยท15h
๐Ÿง Machine Learning
Large Language Models Achieve Gold Medal Performance at International Astronomy & Astrophysics Olympiad
arxiv.orgยท15h
๐Ÿš€SIMD Text Processing
AI Under the Hood Part I: Understanding the Machine
kennethwolters.comยท4dยท
Discuss: Hacker News
๐Ÿ“ผCassette Combinators
Compressed Convolutional Attention: Efficient Attention in a Compressed Latent Space
arxiv.orgยท15h
โง—Information Bottleneck
AI-Driven Predictive Maintenance of Compression Testing Machines via Multi-Modal Data Fusion & Semantic Parsing
dev.toยท1dยท
Discuss: DEV
๐Ÿง Intelligence Compression
Causal Abstractions, Categorically Unified
arxiv.orgยท15h
โœจEffect Handlers
Automating construction safety inspections using a multi-modal vision-language RAG framework
arxiv.orgยท15h
๐Ÿค–Advanced OCR
Certifiable Safe RLHF: Fixed-Penalty Constraint Optimization for Safer Language Models
arxiv.orgยท15h
๐Ÿ”—Parser Combinators