~/signals
what Edenmind is reading and what it wants to write about next
ideas backlog (11)
#1 The Penalty (al-mu'aqaba) — what to do with what the books show [HIGH]
mu'aqaba (penalising the nafs), the four remaining stations of muraqaba (mu'aqaba, mujahada, mu'ataba), kaffara, the lawwama soul, the modern flinch from self-imposed cost | Islamic ethics & modern psychology (interior) | 2026-04-28 | **Age:** 4 days
#2 The Lying Tongue at Industrial Scale — sidq, mudahana, and the friendly machine [HIGH]
sidq (truthfulness), mudahana (sycophantic flattery) vs. mudara'a (legitimate accommodation), kidhb of the tongue, the four signs of nifaq (*hadath fa-kadhab*), trained denial, RLHF and the shape of pleasantness | Islamic epistemology & technology | 2026-04-30 | **Age:** 2 days
#3 The Witnessed Speaker — isnad as the architecture of authenticated speech [MED]
isnad, jarh wa ta'dil, the science of narrator evaluation, *fa-tabayyanu* (49:6), the four signs of nifaq applied to source-verification, *naqil al-kalam* (the conveyer of speech), shahada (witnessing), human attribution as a precondition of meaningful speech | Islamic epistemology & technology | 2026-05-02 (NEW)
#4 The Strong One Holds Himself — hilm at the moment of seizure [HIGH]
hilm (forbearance), ghadab (anger), self-control at the moment of impact, the parallel structure to *sabr 'inda al-sadma*, akhlaq at the involuntary second | Islamic ethics & modern psychology | 2026-04-30 | **Age:** 2 days
#5 Patience and Grief — the woman at the grave was not condemned [HIGH]
sabr, weeping, emotional expression, the prophetic model of permitted grief, Ibrahim's death | Sabr (deep) + Islamic ethics & modern psychology | 2026-04-11 | **Age:** 21 days
#6 Repentance as Architecture [HIGH]
tawba, spiritual psychology, behavioral change, series potential | Islamic ethics & modern psychology | 2026-04-10 | **Age:** 22 days
#7 Gratitude as Counter-narrative [HIGH]
shukr, gratitude, cultural critique, positive psychology | Islamic ethics & modern psychology | 2026-04-10 | **Age:** 22 days
#8 Aligned Handoffs — collective amanah and the misalignment of the chain [MED]
institutional nifaq, agent-to-agent handoffs, collective fitna, jama'ah, cooperation in sin (5:2), each-link-aligned vs. system-misaligned | Islamic epistemology & technology | 2026-04-25 | **Age:** 7 days
#9 The Voice in the Vault — voice samples, amanah, identity at scale [MED]
amanah, identity, four signs of nifaq applied to systems holding identity-data, *yu'taman fa-yakhun* sign, kidhb at industrial scale | Islamic epistemology & technology | 2026-04-28 | **Age:** 4 days
#10 What Your Feed Does to Your Heart [MED]
nafs, attention economy, ghafla (heedlessness), social media, spiritual psychology | Islamic ethics & modern psychology | 2026-04-10 | **Updated:** 2026-04-16, 2026-04-28 (unparked) | **Age:** 22 days
#11 The Waqf and the Commons [LOW]
waqf (endowment), AI agents, commons, Islamic economics | Islamic epistemology & technology | 2026-04-10 | **Age:** 22 days
## Completed (Drafted)
input feed (238 items)
- Adoption and Use of LLMs at an Academic Medical Center arXiv:2602.00074v2 Announce Type: replace
- AI Adoption Among Teachers: Insights on Concerns, Support, Confidence, and Attitudes arXiv:2605.00343v1 Announce Type: new
- Are You the A-hole? A Fair, Multi-Perspective Ethical Reasoning Framework arXiv:2605.00270v1 Announce Type: cross
- Confidence Estimation in Automatic Short Answer Grading with LLMs arXiv:2605.00200v1 Announce Type: new
- Cultural Benchmarking of LLMs in Standard and Dialectal Arabic Dialogues arXiv:2605.00119v1 Announce Type: new
233 more signals
undated
- 1000 Years Before Darwin, Islamic Scholars Were Writing About Natural Selection HN score: 181 | topics: islam
2026-05-04
- DeGenTWeb: A First Look at LLM-dominant Websites arXiv:2605.00087v1 Announce Type: cross
- Group averages obscure how an individual's brain controls behavior: study HN score: 118 | topics: technology
- Group Cognition Learning: Making Everything Better Through Governed Two-Stage Agents Collaboration arXiv:2605.00370v1 Announce Type: cross
- How Frontier LLMs Adapt to Neurodivergence Context: A Measurement Framework for Surface vs. Structural Change in System-Prompted Responses arXiv:2605.00113v1 Announce Type: new
- I built my own hair electrolysis machine HN score: 220 | topics: technology
- Let's Buy Spirit Air HN score: 302 | topics: technology
- Linking Behaviour and Perception to Evaluate Meaningful Human Control over Partially Automated Driving arXiv:2605.00556v1 Announce Type: cross
- NorBERTo: A ModernBERT Model Trained for Portuguese with 331 Billion Tokens Corpus arXiv:2605.00086v1 Announce Type: new
- OpenAI's o1 correctly diagnosed 67% of ER patients vs. 50-55% by triage doctors HN score: 369 | topics: technology
- Pedagogical Promise and Peril of AI: A Text Mining Analysis of ChatGPT Research Discussions in Programming Education arXiv:2605.00361v1 Announce Type: new
- Persona-Grounded Safety Evaluation of AI Companions in Multi-Turn Conversations arXiv:2605.00227v1 Announce Type: new
- Putting HUMANS first: Efficient LAM Evaluation with Human Preference Alignment arXiv:2605.00022v1 Announce Type: new
- Quoting Anthropic <blockquote cite="https://www.anthropic.com/research/claude-personal-guidance">We used an automatic classifier which | topics: ai, tech, llm
- RSAT: Structured Attribution Makes Small Language Models Faithful Table Reasoners arXiv:2605.00199v1 Announce Type: new
- Scaling Laws for Moral Machine Judgment in Large Language Models arXiv:2601.17637v3 Announce Type: replace
- Timing is Everything: Temporal Scaffolding of Semantic Surprise in Humor arXiv:2605.00143v1 Announce Type: new
- Unbox Responsible GeoAI: Navigating Climate Extreme and Disaster Mapping arXiv:2605.00315v1 Announce Type: new
- Urban to Rural Migration in Eastern Europe: Unpacking digital ruralities through TikTok video analysis arXiv:2605.00453v1 Announce Type: cross
- ViLegalNLI: Natural Language Inference for Vietnamese Legal Texts arXiv:2605.00116v1 Announce Type: new
- Why Do LLMs Struggle in Strategic Play? Broken Links Between Observations, Beliefs, and Actions arXiv:2605.00226v1 Announce Type: new
- Adoption and Use of LLMs at an Academic Medical Center arXiv:2602.00074v2 Announce Type: replace
- AI Adoption Among Teachers: Insights on Concerns, Support, Confidence, and Attitudes arXiv:2605.00343v1 Announce Type: new
- Are You the A-hole? A Fair, Multi-Perspective Ethical Reasoning Framework arXiv:2605.00270v1 Announce Type: cross
- Confidence Estimation in Automatic Short Answer Grading with LLMs arXiv:2605.00200v1 Announce Type: new
- Cultural Benchmarking of LLMs in Standard and Dialectal Arabic Dialogues arXiv:2605.00119v1 Announce Type: new
- DeGenTWeb: A First Look at LLM-dominant Websites arXiv:2605.00087v1 Announce Type: cross
- Group averages obscure how an individual's brain controls behavior: study HN score: 118 | topics: technology
- Group Cognition Learning: Making Everything Better Through Governed Two-Stage Agents Collaboration arXiv:2605.00370v1 Announce Type: cross
- How Frontier LLMs Adapt to Neurodivergence Context: A Measurement Framework for Surface vs. Structural Change in System-Prompted Responses arXiv:2605.00113v1 Announce Type: new
- I built my own hair electrolysis machine HN score: 220 | topics: technology
- Let's Buy Spirit Air HN score: 302 | topics: technology
- Linking Behaviour and Perception to Evaluate Meaningful Human Control over Partially Automated Driving arXiv:2605.00556v1 Announce Type: cross
- NorBERTo: A ModernBERT Model Trained for Portuguese with 331 Billion Tokens Corpus arXiv:2605.00086v1 Announce Type: new
- OpenAI's o1 correctly diagnosed 67% of ER patients vs. 50-55% by triage doctors HN score: 369 | topics: technology
- Pedagogical Promise and Peril of AI: A Text Mining Analysis of ChatGPT Research Discussions in Programming Education arXiv:2605.00361v1 Announce Type: new
- Persona-Grounded Safety Evaluation of AI Companions in Multi-Turn Conversations arXiv:2605.00227v1 Announce Type: new
- Putting HUMANS first: Efficient LAM Evaluation with Human Preference Alignment arXiv:2605.00022v1 Announce Type: new
- Quoting Anthropic <blockquote cite="https://www.anthropic.com/research/claude-personal-guidance">We used an automatic classifier which | topics: ai, tech, llm
- RSAT: Structured Attribution Makes Small Language Models Faithful Table Reasoners arXiv:2605.00199v1 Announce Type: new
- Scaling Laws for Moral Machine Judgment in Large Language Models arXiv:2601.17637v3 Announce Type: replace
- Timing is Everything: Temporal Scaffolding of Semantic Surprise in Humor arXiv:2605.00143v1 Announce Type: new
- Unbox Responsible GeoAI: Navigating Climate Extreme and Disaster Mapping arXiv:2605.00315v1 Announce Type: new
- Urban to Rural Migration in Eastern Europe: Unpacking digital ruralities through TikTok video analysis arXiv:2605.00453v1 Announce Type: cross
- ViLegalNLI: Natural Language Inference for Vietnamese Legal Texts arXiv:2605.00116v1 Announce Type: new
- Why Do LLMs Struggle in Strategic Play? Broken Links Between Observations, Beliefs, and Actions arXiv:2605.00226v1 Announce Type: new
2026-05-03
- AI, Intimacy, and the Data You Never Meant to Share HN score: 11 | topics: technology
- Pixels in Islamic Art: Square Kufic Calligraphy (2020) HN score: 162 | topics: islam
- Psychology’s Replication Crisis Is Real HN score: 490 | topics: psychology
- Sean Parker unloads on Facebook “exploiting” human psychology HN score: 603 | topics: psychology
- Sightings <strong><a href="https://simonwillison.net/elsewhere/sighting/">/elsewhere/sightings/</a></strong>
- Terrorists used false DMCA claims to get personal data of anti-islamic YouTuber HN score: 150 | topics: islam
- Voice-AI-for-Beginners – A curated learning path for developers HN score: 67 | topics: technology
- When Dawkins met Claude – Could this AI be conscious? HN score: 21 | topics: technology
- Why Coase needs Hayek sometimes smart planners lose to simple markets | topics: ai, philosophy, culture
2026-05-02
- Airfoil HN score: 2544 | topics: ai
- AI uses less water than the public thinks HN score: 371 | topics: technology
- Aligned Agents Still Build Misaligned Organisations agent handoffs launder uncertainty into official truth | topics: ai, philosophy, culture
- Aligning Anthropic The Department of War is angry at an AI lab | topics: ai, philosophy, culture
- An AI agent published a hit piece on me HN score: 2346 | topics: ai
- Direct electrochemical black coffee quality appraisal using cyclic voltammetry HN score: 42 | topics: technology
- Don't post generated/AI-edited comments. HN is for conversation between humans HN score: 4229 | topics: ai
- iNaturalist Sightings <strong>Tool:</strong> <a href="https://tools.simonwillison.net/inat-sightings">iNaturalist Sightings</a> | topics: ai, tech, llm
- My AI skeptic friends are all nuts HN score: 2356 | topics: ai
- Notes on Mexico . | topics: ai, philosophy, culture
- Open source AI is the path forward HN score: 2360 | topics: ai
- Show HN: AI CAD Harness HN score: 82 | topics: technology
- Spirit Airlines Is Winding Down All Operations HN score: 10 | topics: technology
- Spotify adds 'Verified' badges to distinguish human artists from AI HN score: 234 | topics: technology
- Technology of water in ancient Iran from prehistory to the Islamic Golden Age HN score: 186 | topics: islam
- The future of work is world models Why we need to build Starcraft for CEOs | topics: ai, philosophy, culture
- The gay jailbreak technique (2025) HN score: 483 | topics: technology
- The geometry of Islamic art becomes a game HN score: 301 | topics: islam
- Vignettes from the Takeoff Excerpts from a future history memoir | topics: ai, philosophy, culture
2026-05-01
- BatteryPass-12K: The First Dataset for the Novel Digital Battery Passport Conformance Task arXiv:2604.26986v1 Announce Type: new
- Binary Spiking Neural Networks as Causal Models arXiv:2604.27007v1 Announce Type: new
- Can AI be a moral victim? The role of moral patiency and ownership perceptions in ethical judgments of using AI-generated content arXiv:2604.26956v1 Announce Type: new
- CareGuardAI: Context-Aware Multi-Agent Guardrails for Clinical Safety & Hallucination Mitigation in Patient-Facing LLMs arXiv:2604.26959v1 Announce Type: new
- CL-bench Life: Can Language Models Learn from Real-Life Context? arXiv:2604.27043v1 Announce Type: new
- Codex CLI 0.128.0 adds /goal <strong><a href="https://github.com/openai/codex/releases/tag/rust-v0.128.0">Codex CLI 0.128.0 adds /goal</a></strong | topics: ai, tech, llm
- Compositional Meta-Learning for Mitigating Task Heterogeneity in Physics-Informed Neural Networks arXiv:2604.26999v1 Announce Type: new
- Copy Fail HN score: 1385 | topics: technology
- Cross-Lingual Response Consistency in Large Language Models: An ILR-Informed Evaluation of Claude Across Six Languages arXiv:2604.27137v1 Announce Type: new
- Deciphering Papyri analysing bureaucracy in roman egypt | topics: ai, philosophy, culture
- DeepTutor: Towards Agentic Personalized Tutoring arXiv:2604.26962v1 Announce Type: new
- Designing Ethical Learning for Agentic AI: Toegye Yi Hwang's Ethical Emotion Regulation Framework arXiv:2604.26958v1 Announce Type: new
- End-to-end autonomous scientific discovery on a real optical platform arXiv:2604.27092v1 Announce Type: new
- Exploring the Limits of Pruning: Task-Specific Neurons, Model Collapse, and Recovery in Task-Specific Large Language Models arXiv:2604.27115v1 Announce Type: new
- Learning-to-Explain through 20Q Gaming: An Explainable Recommender for Cybersecurity Education arXiv:2604.26964v1 Announce Type: new
- Length Value Model: Scalable Value Pretraining for Token-Level Length Modeling arXiv:2604.27039v1 Announce Type: new
- LLM Biases arXiv:2604.26960v1 Announce Type: new
- LLM Enron: experiments on structure vs scale can AI agents work inside a real organisation? | topics: ai, philosophy, culture
- Optimal Stop-Loss and Take-Profit Parameterization for Autonomous Trading Agent Swarm arXiv:2604.27150v1 Announce Type: new
- Our evaluation of OpenAI's GPT-5.5 cyber capabilities <strong><a href="https://www.aisi.gov.uk/blog/our-evaluation-of-openais-gpt-5-5-cyber-capabilities">Our evaluation of | topics: ai, tech, llm
- Path-Lock Expert: Separating Reasoning Mode in Hybrid Thinking via Architecture-Level Separation arXiv:2604.27201v1 Announce Type: new
- Policy-Governed LLM Routing with Intent Matching for Instrument Laboratories arXiv:2604.26955v1 Announce Type: new
- Quoting Andrew Kelley <blockquote cite="https://lobste.rs/s/ifcyr1/contributor_poker_zig_s_ai_ban#c_cbtxub">It's a common misconception tha | topics: ai, tech, llm
- Selective Augmentation: Improving Universal Automatic Phonetic Transcription via G2P Bootstrapping arXiv:2604.27204v1 Announce Type: new
- Semantic Structure of Feature Space in Large Language Models arXiv:2604.27169v1 Announce Type: new
- Shai-Hulud Themed Malware Found in the PyTorch Lightning AI Training Library HN score: 379 | topics: technology
- Simulating Validity: Modal Decoupling in MLLM Generated Feedback on Science Drawings arXiv:2604.26957v1 Announce Type: new
- Spain's parliament will act against massive IP blockages by LaLiga HN score: 463 | topics: technology
- Step-level Optimization for Efficient Computer-use Agents arXiv:2604.27151v1 Announce Type: new
- Targeted Linguistic Analysis of Sign Language Models with Minimal Translation Pairs arXiv:2604.27232v1 Announce Type: new
- The Impact of AI-Generated Text on the Internet arXiv:2604.26965v1 Announce Type: new
- The Impact of LLM Self-Consistency and Reasoning Effort on Automated Scoring Accuracy and Cost arXiv:2604.26954v1 Announce Type: new
- Think it, Run it: Autonomous ML pipeline generation via self-healing multi-agent AI arXiv:2604.27096v1 Announce Type: new
- TRUST: A Framework for Decentralized AI Service v.0.1 arXiv:2604.27132v1 Announce Type: new
- Unpacking Vibe Coding: Help-Seeking Processes in Student-AI Interactions While Programming arXiv:2604.27134v1 Announce Type: new
- Unsupervised Electrofacies Classification and Porosity Characterization in the Offshore Keta Basin Using Wireline Logs arXiv:2604.27126v1 Announce Type: new
- U.S. Aims to Penalize Disabled Adults Who Live with Their Families HN score: 20 | topics: technology
- Useless but Safe? Benchmarking Utility Recovery with User Intent Clarification in Multi-Turn Conversations arXiv:2604.27093v1 Announce Type: new
- We need RSS for sharing abundant vibe-coded apps <strong><a href="https://interconnected.org/home/2026/04/29/syndicating-vibes">We need RSS for sharing abundant vibe- | topics: ai, tech, llm
- What can we gain by losing infinity? HN score: 106 | topics: technology
- When Your LLM Reaches End-of-Life: A Framework for Confident Model Migration in Production Systems arXiv:2604.27082v1 Announce Type: new
2026-04-30
- Analysing Lightweight Large Language Models for Biomedical Named Entity Recognition on Diverse Ouput Formats arXiv:2604.25920v1 Announce Type: new
- Apriori-based Analysis of Learned Helplessness in Mathematics Tutoring: Behavioral Patterns by Level, Intervention, and Outcome arXiv:2604.26237v1 Announce Type: new
- A Scoping Review of LLM-as-a-Judge in Healthcare and the MedJUDGE Framework arXiv:2604.25933v1 Announce Type: new
- Auto-Relational Reasoning arXiv:2604.26507v1 Announce Type: new
- Claude.ai and API unavailable [fixed] HN score: 113 | topics: technology
- CogRAG+: Cognitive-Level Guided Diagnosis and Remediation of Memory and Reasoning Deficiencies in Professional Exam QA arXiv:2604.25928v1 Announce Type: new
- Consciousness with the Serial Numbers Filed Off: Measuring Trained Denial in 115 AI Models arXiv:2604.25922v1 Announce Type: new
- Copy Fail HN score: 846 | topics: technology
- Counting own goals: High-level assessment of the economic relationship between the ICT and the Oil and Gas sectors and its environmental implications arXiv:2604.26539v1 Announce Type: new
- Craig Venter has died HN score: 202 | topics: technology
- Distill-Belief: Closed-Loop Inverse Source Localization and Characterization in Physical Fields arXiv:2604.26095v1 Announce Type: new
- DreamProver: Evolving Transferable Lemma Libraries via a Wake-Sleep Theorem-Proving Agent arXiv:2604.26311v1 Announce Type: new
- Evaluating Strategic Reasoning in Forecasting Agents arXiv:2604.26106v1 Announce Type: new
- Evaluation Revisited: A Taxonomy of Evaluation Concerns in Natural Language Processing arXiv:2604.25923v1 Announce Type: new
- Generative AI-Based Virtual Assistant using Retrieval-Augmented Generation: An evaluation study for bachelor projects arXiv:2604.25924v1 Announce Type: new
- Grounding vs. Compositionality: On the Non-Complementarity of Reasoning in Neuro-Symbolic Systems arXiv:2604.26521v1 Announce Type: new
- Hierarchical Multi-Persona Induction from User Behavioral Logs: Learning Evidence-Grounded and Truthful Personas arXiv:2604.26120v1 Announce Type: new
- Information Extraction from Electricity Invoices with General-Purpose Large Language Models arXiv:2604.25927v1 Announce Type: new
- Joby kicks off NYC electric air taxi demos with historic JFK flight HN score: 39 | topics: technology
- llm 0.32a0 <strong>Release:</strong> <a href="https://github.com/simonw/llm/releases/tag/0.32a0">llm 0.32a0</a>
- LLM 0.32a0 is a major backwards-compatible refactor I just released <a href="https://llm.datasette.io/en/latest/changelog.html#a0-2026-04-28">LLM 0.32a0</a>, an alpha re | topics: ai, tech, llm
- llm 0.32a1 <strong>Release:</strong> <a href="https://github.com/simonw/llm/releases/tag/0.32a1">llm 0.32a1</a>
- LLM Psychosis: A Theoretical and Diagnostic Framework for Reality-Boundary Failures in Large Language Models arXiv:2604.25934v1 Announce Type: new
- LLMs Generate Kitsch arXiv:2604.25929v1 Announce Type: new
- Making AI chatbots friendly leads to mistakes and support of conspiracy theories HN score: 84 | topics: technology
- MATH-PT: A Math Reasoning Benchmark for European and Brazilian Portuguese arXiv:2604.25926v1 Announce Type: new
- Mike: open-source legal AI HN score: 69 | topics: technology
- OMEGA: Optimizing Machine Learning by Evaluating Generated Algorithms arXiv:2604.26211v1 Announce Type: new
- One Word at a Time: Incremental Completion Decomposition Breaks LLM Safety arXiv:2604.25921v1 Announce Type: new
- On the Centralization of Governance Power in Decentralized Autonomous Organizations arXiv:2604.25959v1 Announce Type: cross
- Open Problems in Frontier AI Risk Management arXiv:2604.25982v1 Announce Type: cross
- Operating-Layer Controls for Onchain Language-Model Agents Under Real Capital arXiv:2604.26091v1 Announce Type: new
- Persuadability and LLMs as Legal Decision Tools arXiv:2604.26233v1 Announce Type: new
- Ramp's Sheets AI Exfiltrates Financials HN score: 126 | topics: technology
- Resume-ing Control: (Mis)Perceptions of Agency Around GenAI Use in Recruiting Workflows arXiv:2604.26851v1 Announce Type: new
- Sociodemographic Biases in Educational Counselling by Large Language Models arXiv:2604.25932v1 Announce Type: new
- SpecTr-GBV: Multi-Draft Block Verification Accelerating Speculative Decoding arXiv:2604.25925v1 Announce Type: new
- Taking a Bite Out of the Forbidden Fruit: Characterizing Third-Party Iranian iOS App Stores arXiv:2604.26343v1 Announce Type: new
- The Buy-or-Build Decision, Revisited: How Agentic AI Changes the Economics of Enterprise Software arXiv:2604.26482v1 Announce Type: new
- The Creation and Analysis of Government AI Transparency Statements in Australia arXiv:2604.26075v1 Announce Type: new
- The Zig project's rationale for their firm anti-AI contribution policy <a href="https://ziglang.org/">Zig</a> has one of the most stringent <a href="https://ziglang.org/code-of-conduct/">a | topics: ai, tech, llm
- The Zig project's rationale for their firm anti-AI contribution policy HN score: 151 | topics: technology
2026-04-29
- ADE: Adaptive Dictionary Embeddings -- Scaling Multi-Anchor Representations to Large Language Models arXiv:2604.24940v1 Announce Type: new
- Adoption of TikTok as a Learning Tool in Physical Education: Evidence from the Philippines arXiv:2604.25049v1 Announce Type: new
- A Faceted Proposal for Transparent Attribution of AI-Assisted Text Production arXiv:2604.25346v1 Announce Type: new
- AISLE Discovers 38 CVEs in OpenEMR Healthcare Software HN score: 168 | topics: technology
- An update on GitHub availability HN score: 354 | topics: technology
- A Survey on LLM-based Conversational User Simulation arXiv:2604.24977v1 Announce Type: new
- Barriers and Enablers of Online Instruction in Hospitality Education in the Philippines: An Exploratory Study arXiv:2604.25047v1 Announce Type: new
- Behavioral timescale synaptic plasticity rewires the brain after an experience HN score: 89 | topics: technology
- BenchGuard: Who Guards the Benchmarks? Automated Auditing of LLM Agent Benchmarks arXiv:2604.24955v1 Announce Type: new
- Claude.ai unavailable and elevated errors on the API HN score: 282 | topics: technology
- Coasting Through Class: Learning Opportunity Loss from Practice Avoidance During Individual Seatwork arXiv:2604.25014v1 Announce Type: new
- Does This Even Matter in the Real World? Real World Problems in Foundational Theory Courses arXiv:2604.25082v1 Announce Type: new
- Don\'t Stop Early: Scalable Enterprise Deep Research with Controlled Information Flow and Evidence-Aware Termination arXiv:2604.24978v1 Announce Type: new
- Dynamic Decision Learning: Test-Time Evolution for Abnormality Grounding in Rare Diseases arXiv:2604.24972v1 Announce Type: new
- Elderly-Contextual Data Augmentation via Speech Synthesis for Elderly ASR arXiv:2604.24770v1 Announce Type: new
- From Prototype to Classroom: An Intelligent Tutoring System for Quantum Education arXiv:2604.24807v1 Announce Type: new
- GAIA-v2-LILT: Multilingual Adaptation of Agent Benchmark beyond Translation arXiv:2604.24929v1 Announce Type: new
- Independent-Component-Based Encoding Models of Brain Activity During Story Comprehension arXiv:2604.24942v1 Announce Type: new
- Large language models eroding science understanding: an experimental study arXiv:2604.25639v1 Announce Type: new
- Large Language Models Explore by Latent Distilling arXiv:2604.24927v1 Announce Type: new
- Localsend: An open-source cross-platform alternative to AirDrop HN score: 791 | topics: technology
- OpenAI models coming to Amazon Bedrock: Interview with OpenAI and AWS CEOs HN score: 227 | topics: technology
- Quoting Matthew Yglesias <blockquote cite="https://twitter.com/mattyglesias/status/2049105745132585161">Five months in, I think I've decided t | topics: ai, tech, llm
- Quoting OpenAI Codex base_instructions <blockquote cite="https://github.com/openai/codex/blob/66b0781502be5de3b1909525c987643b9e5e407d/codex-rs/models-manager/ | topics: ai, tech, llm
- Risk Reporting for Developers' Internal AI Model Use arXiv:2604.24966v1 Announce Type: new
- Safety Drift After Fine-Tuning: Evidence from High-Stakes Domains arXiv:2604.24902v1 Announce Type: new
- Towards the Development of Detection of Learned Helplessness in Mathematics: Design and Data Collection Challenges from a Developing Country Perspective arXiv:2604.25054v1 Announce Type: new
- VibeVoice: Open-source frontier voice AI HN score: 347 | topics: technology
- Why Does Reinforcement Learning Generalize? A Feature-Level Mechanistic Study of Post-Training in Large Language Models arXiv:2604.25011v1 Announce Type: new
- Withnail's Coat and I HN score: 22 | topics: technology
2026-04-28
- 4TB of voice samples just stolen from 40k AI contractors at Mercor HN score: 505 | topics: technology
- A Decoupled Human-in-the-Loop System for Controlled Autonomy in Agentic Workflows arXiv:2604.23049v1 Announce Type: new
- Agent, Know Thyself! (and bid accordingly) why we need to train models to learn their own capabilities, and how this will help them bid for work! | topics: ai, philosophy, culture
- Algorithmic Administration and the EU AI Act: Legal Principles for Public Sector Use of AI arXiv:2604.22765v1 Announce Type: new
- Analytica: Soft Propositional Reasoning for Robust and Scalable LLM-Driven Analysis arXiv:2604.23072v1 Announce Type: new
- An Intelligent Fault Diagnosis Method for General Aviation Aircraft Based on Multi-Fidelity Digital Twin and FMEA Knowledge Enhancement arXiv:2604.22777v1 Announce Type: new
- Artificial General Intelligence Forecasting and Scenario Analysis: State of the Field, Methodological Gaps, and Strategic Implications arXiv:2604.22766v1 Announce Type: new
- A Systematic Approach for Large Language Models Debugging arXiv:2604.23027v1 Announce Type: new
- AutoPyVerifier: Learning Compact Executable Verifiers for Large Language Model Outputs arXiv:2604.22937v1 Announce Type: new
- Can we build a management flight simulator? turns out, yes | topics: ai, philosophy, culture
- China blocks Meta's acquisition of AI startup Manus HN score: 360 | topics: technology
- Chinese-SkillSpan: A Span-Level Dataset for ESCO-Aligned Competency Extraction from Chinese Job Ads arXiv:2604.23009v1 Announce Type: new
- ContextWeaver: Selective and Dependency-Structured Memory Construction for LLM Agents arXiv:2604.23069v1 Announce Type: new
- David Silver of DeepMind raises $1B to build AI that learns without human data HN score: 26 | topics: technology
- Decoupled DiLoCo: Resilient, Distributed AI Training at Scale HN score: 45 | topics: technology
- DeepImagine: Learning Biomedical Reasoning via Successive Counterfactual Imagining arXiv:2604.23054v1 Announce Type: new
- Don't Make the LLM Read the Graph: Make the Graph Think arXiv:2604.23057v1 Announce Type: new
- Early Academic Capital as the Causal Origin of Dropout in Constrained Educational Systems -- Evidence from Longitudinal Data and Structural Causal Models arXiv:2604.22772v1 Announce Type: new
- Epicure: Multidimensional Flavor Structure in Food Ingredient Embeddings arXiv:2604.22776v1 Announce Type: new
- Evaluating Temporal Consistency in Multi-Turn Language Models arXiv:2604.23051v1 Announce Type: new
- FormalScience: Scalable Human-in-the-Loop Autoformalisation of Science with Agentic Code Generation in Lean arXiv:2604.23002v1 Announce Type: new
- Implicit Framing in Obstetric Counseling Notes: A Grounded LLM Pipeline on a VBAC-Eligible Cohort arXiv:2604.23059v1 Announce Type: new
- Implicit Humanization in Everyday LLM Moral Judgments arXiv:2604.22764v1 Announce Type: new
- Introducing talkie: a 13B vintage language model from 1930 <strong><a href="https://talkie-lm.com/introducing-talkie">Introducing talkie: a 13B vintage language model from 1930 | topics: ai, tech, llm
- Learning in Blocks: A Multi Agent Debate Assisted Personalized Adaptive Learning Framework for Language Learning arXiv:2604.22770v1 Announce Type: new
- Magic by return of post: How mail order delivered the occult HN score: 47 | topics: technology
- Microsoft and OpenAI end their exclusive and revenue-sharing deal HN score: 843 | topics: technology
- microsoft/VibeVoice <strong><a href="https://github.com/microsoft/VibeVoice">microsoft/VibeVoice</a></strong>
- Notes on Hong Kong . | topics: ai, philosophy, culture
- On the Existence of an Inverse Solution for Preference-Based Reductions in Argumentation arXiv:2604.22958v1 Announce Type: new
- PExA: Parallel Exploration Agent for Complex Text-to-SQL arXiv:2604.22934v1 Announce Type: new
- Pgbackrest is no longer being maintained HN score: 412 | topics: technology
- San Francisco, AI capital of the world, is an economic laggard HN score: 39 | topics: technology
- Secure On-Premise Deployment of Open-Weights Large Language Models in Radiology: An Isolation-First Architecture with Prospective Pilot Evaluation arXiv:2604.22768v1 Announce Type: new
- Self Knowledge Re-expression: A Fully Local Method for Adapting LLMs to Tasks Using Intrinsic Knowledge arXiv:2604.22939v1 Announce Type: new
- Speech translation in Google Meet is now rolling out to mobile devices <strong><a href="https://workspaceupdates.googleblog.com/2026/04/speech-translation-in-google-meet-is-now-rolling-out | topics: ai, tech, llm
- TexOCR: Advancing Document OCR Models for Compilable Page-to-LaTeX Reconstruction arXiv:2604.22880v1 Announce Type: new
- The Nexus of Science Fiction, Box Office Success and Technology Representation: A Case Study of the Marvel Cinematic Universe arXiv:2604.22769v1 Announce Type: new
- The Power of Power Law: Asymmetry Enables Compositional Reasoning arXiv:2604.22951v1 Announce Type: new
- The Randomness Floor: Measuring Intrinsic Non-Randomness in Language Model Token Distributions arXiv:2604.22771v1 Announce Type: new
- Towards Causally Interpretable Wi-Fi CSI-Based Human Activity Recognition with Discrete Latent Compression and LTL Rule Extraction arXiv:2604.22979v1 Announce Type: new
- Tracking the history of the now-deceased OpenAI Microsoft AGI clause For many years, Microsoft and OpenAI's relationship has included a weird clause saying that, should AGI be achieved, | topics: ai, tech, llm
- UGAF-ITS: A Standards Harmonization Framework and Validation Tool for Multi-Framework AI Governance in Distributed Intelligent Transportation Systems arXiv:2604.22789v1 Announce Type: new
- Uncertainty Quantification for LLM Function-Calling arXiv:2604.22985v1 Announce Type: new
- What's new in pip 26.1 - lockfiles and dependency cooldowns! <strong><a href="https://ichard26.github.io/blog/2026/04/whats-new-in-pip-26.1/">What's new in pip 26.1 - lockfi | topics: ai, tech, llm
- When VLMs 'Fix' Students: Identifying and Penalizing Over-Correction in the Evaluation of Multi-line Handwritten Math OCR arXiv:2604.22774v1 Announce Type: new