News
The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.
| Source | Age | Title | Summary | Status |
|---|---|---|---|---|
arxiv arxiv-ai | 1mo | TRACE: Accountable Agentic Retrieval for Source Discovery in Digital Archives | Historical archives pose a difficult retrieval problem for retrievalaugmented generation systems: documents are OCR-degraded, heteroge… | ↗Not briefed |
hn hn-ai | 1mo | Ask HN: How do you use AWS AssumeRole? | i want to using short-lived sessions with aws console login credentials (already support on awscli)<p>i have been using Leapp so i tri… | ↗Not briefed |
arxiv arxiv-ai | 1mo | ClashBench: Conflicts Leading Agents to Seize and Harm | As agent systems become more widely used, multiple agent sessions increasingly run alongside pre-existing user tasks in the same envir… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Online Adaptive Kernel Mixing for Gaussian Process Decision Making | Gaussian Processes (GPs) are widely used as surrogates for black-box functions in sequential decision-making problems such as Bayesian… | ↗Not briefed |
hn hn-ai | 1mo | Reckless Expansion Plans Blow Past European Airports Carbon Limits | — | ↗Not briefed |
hn hn-ai | 1mo | AI Safety Is Mostly a Sex Cult | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Jämförelsen håller helt ok om man ser de kommersiella LLM-produkterna som de gigantiska SUV-ar som amerikanerna kör runt med. Slukar energ… | Jämförelsen håller helt ok om man ser de kommersiella LLM-produkterna som de gigantiska SUV-ar som amerikanerna kör runt med. Slukar… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Generalization through Lexical Abstraction in Transformer Models: The Case of Functional Words | Pronouns, adverbs and other functional words (such as they, her, somewhere, there) are often used in language to replace concrete noun… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI admite otros seis incidentes alarmantes con sus modelos que ignoraban las reglas de sus creadores | OpenAI admite otros seis incidentes alarmantes con sus modelos que ignoraban las reglas de sus creadores | ↗Not briefed |
arxiv arxiv-ai | 1mo | Evaluating Communicative Success in Machine-Translated Conversation | Interpreter agents built on machine translation (MT) increasingly mediate live conversation between people who do not share a language… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Un programa no farà mai res q no s'hagi programat. Sigui verbalment, sigui amb codi. Continuen jugant a "la IA se m'ha escapat" quan és evi… | Un programa no farà mai res q no s'hagi programat. Sigui verbalment, sigui amb codi. Continuen jugant a "la IA se m'ha escapat" quan é… | ↗Not briefed |
arxiv arxiv-ai | 1mo | PetriBench: Benchmarking LLM Reasoning over Dynamic State Spaces | Characterizing LLM reasoning remains an open challenge, as many existing benchmarks isolate specific reasoning skills, rely on externa… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Der ChatGPT-Entwickler OpenAI hat sechs neue Zwischenfälle öffentlich gemacht, bei denen sich seine Künstliche Intelligenz (KI) in Tests "…🔧 ChatGPT | "Der ChatGPT-Entwickler OpenAI hat sechs neue Zwischenfälle öffentlich gemacht, bei denen sich seine Künstliche Intelligenz (KI) in Te… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Et veldig tydelig resultat av antropomorfiseringen av "AI" er at selskapene kan snakke om denne typen hendelser som "noe AI gjør", og ikke … | Et veldig tydelig resultat av antropomorfiseringen av "AI" er at selskapene kan snakke om denne typen hendelser som "noe AI gjør", og… | ↗Not briefed |
arxiv arxiv-ai | 1mo | D-Quant: Driftable Entropy Coding for KV Cache Quantization | The KV cache has become a major bottleneck in deploying LLMs, as its memory footprint grows linearly with sequence length and batch si… | ↗Not briefed |
arxiv arxiv-ai | 1mo | VākQA: A Benchmark and Evaluation Study for Telugu Spoken Factoid Question Answering | Question answering has advanced rapidly with large language models, but predominantly for high-resource languages, in both text and sp… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Uni-LaDiR: Latent Diffusion Unifies Multimodal Reasoning | Multimodal reasoning requires models to draw on information from multiple modalities throughout the reasoning process. Yet existing me… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 🤖 Cando a Intelixencia Artificial intenta facer trampas: OpenAI destapa seis novos incidentes cos seus modelos www.nosdiario.gal/articulo… | 🤖 Cando a Intelixencia Artificial intenta facer trampas: OpenAI destapa seis novos incidentes cos seus modelos www.nosdiario.gal/art… | ↗Not briefed |
arxiv arxiv-ai | 1mo | JustMem: Just-Enough Memory Access for Long-Term Conversations | Efficient long-term conversational memory requires retrieving sufficient evidence without indiscriminately expanding the context prese… | ↗Not briefed |
hn hn-ai | 1mo | GLM Built Its Own Inference Infrastructure | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | En LLM kan inte ge råd, utan bara sätta ihop de mest logiskt följande orden utifrån en stor databank. Den som använder en LLM lär sig ingen… | En LLM kan inte ge råd, utan bara sätta ihop de mest logiskt följande orden utifrån en stor databank. Den som använder en LLM lär sig… | ↗Not briefed |
arxiv arxiv-ai | 1mo | AURA: Adaptive Uncertainty-Routed Analysis for Email Threat Detection | Email spam and phishing attacks remain a critical security threat. Adversaries increasingly exploit large language models to craft con… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Physical knowledge on historical data matters more than enforcing physical constraints on the forecast | Time series forecasting has seen signicant advancements with the emergence of new deep learning models. However, forecasting time seri… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Microsoft's AI boss: don't anthropomorphise AI. Also him, same article: we're "seeding a new silicon species which will no doubt compete w… | Microsoft's AI boss: don't anthropomorphise AI. Also him, same article: we're "seeding a new silicon species which will no doubt comp… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Zarya: A Hybrid Autoregressive--Masked Diffusion Language Model with Flexible Training and Dual-Mode Inference | Autoregressive language models (ARMs) are constrained by sequential, left-to-right generation, while masked diffusion models (MDMs) en… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Reproducibility is not construct validity: LLM measurement of institutionally situated communication | High annotation reproducibility does not necessarily imply that an LLM-inferred measure captures the construct it is intended to measu… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Pretrained Medical Representations for the Practical Screening of Drug Repositioning Candidates | Representation learning from medical code sequences in electronic health records and medical claims data has been successful in variou… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic Merges Claude Chat and Cowork in One Hub thailandai.news/anthropic-me... #ThailandAINews #ThailandAI #AINews #News #Thailand #B… | Anthropic Merges Claude Chat and Cowork in One Hub thailandai.news/anthropic-me... #ThailandAINews #ThailandAI #AINews #News #Thaila… | ↗Not briefed |
hn hn-ai | 1mo | (Go) Spelunk | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | 本当にこういう者が占い師名乗ったり祈祷師名乗ったりするの辞めていただきたい | 本当にこういう者が占い師名乗ったり祈祷師名乗ったりするの辞めていただきたい | ↗Not briefed |
hn hn-ai | 1mo | EpicSemi Contrail AIX RISC-V Servers | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Expected Hypervolume Maximization for Multiobjective Optimization under Uncertainties | The problem of multiobjective optimization under uncertainties is often approached by taking the expectation of each objective. In thi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | ちょっと前まではネットのデータなんて玉石混交で自分で真贋を見極めなければいけない、って分かってたはずの人達が 「AI(現状のはLLMなので人工知能でさえない)は集合知」 とか大真面目に連呼してたのもうヤバすぎて笑えなかった。 もはや宗教と変わらんよこんなの。 | ちょっと前まではネットのデータなんて玉石混交で自分で真贋を見極めなければいけない、って分かってたはずの人達が 「AI(現状のはLLMなので人工知能でさえない)は集合知」 とか大真面目に連呼してたのもうヤバすぎて笑えなかった。 もはや宗教と変わらんよこんなの。 | ↗Not briefed |
hn hn-ai | 1mo | Measuring and Exploiting Implicit Trust in LLM Tool-Calling Pipelines | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | A Functional Pilot for Certified Freshness-Aware Semantic--Spatial Range Retrieval | Geographic applications need every object inside a radius that satisfies a semantic threshold, yet embedding indexes return approximat… | ↗Not briefed |
bluesky bluesky-ai | 2mo | KI-Firmen warnen in drastischer Rhetorik vor einem durch die eigenen Produkte verursachten Weltuntergang. Ich kommentiere bei @riffreporter… | KI-Firmen warnen in drastischer Rhetorik vor einem durch die eigenen Produkte verursachten Weltuntergang. Ich kommentiere bei @riffrep… | ↗Not briefed |
arxiv arxiv-ai | 2mo | PACE: Precise AI Cinematic Expression: A Typed Specification for Script-Grounded Previsualization and Geometric Conformance | Between a screenplay and a film sits a planning problem that is spatial first: who stands where, and what a camera sees from where it… | ↗Not briefed |
bluesky bluesky-ai | 2mo | After a short experiment with a local LLM, I’ve switched glassbead.games back to its original NLP approach. #AIFree! And I just added Blue… | After a short experiment with a local LLM, I’ve switched glassbead.games back to its original NLP approach. #AIFree! And I just added… | ↗Not briefed |
bluesky bluesky-ai | 2mo | 名誉のために言っておくと同僚はみんなベテラン技術者でワイなんかより頭良かったはずの人達。 それがそろいもそろってLLMに関してだけはお見事な理解を示している。 イライザ効果で人工知能だと思いたい感情が知識や思考より優先している。 | 名誉のために言っておくと同僚はみんなベテラン技術者でワイなんかより頭良かったはずの人達。 それがそろいもそろってLLMに関してだけはお見事な理解を示している。 イライザ効果で人工知能だと思いたい感情が知識や思考より優先している。 | ↗Not briefed |
rss lastweekin-ai | 2mo | Last Week in AI #344 - Navier–Stokes, Pacing the Frontier, AI Misuse | OpenAI claims a Millennium Prize proof amid a feud with mathematicians, Anthropic's CEO calls to pace the frontier, extinction warning… | ↗Not briefed |
bluesky bluesky-ai | 2mo | I’m not sure how widely this is being reported, continuing OpenAI offsec AI fallout. eu.tennessean.com/story/news/e... | I’m not sure how widely this is being reported, continuing OpenAI offsec AI fallout. eu.tennessean.com/story/news/e... | ↗Not briefed |
bluesky bluesky-ai | 2mo | I have tried telling my LLM to create a billion dollar business quick, but that hasn't really worked well so far. Maybe I ought to try addi… | I have tried telling my LLM to create a billion dollar business quick, but that hasn't really worked well so far. Maybe I ought to try… | ↗Not briefed |
bluesky bluesky-ai | 2mo | 研修で同僚がLLMのことを人工知能って言い張ってる時に「AIは集合知だから!」って何度も言っててめっちゃ笑うのを我慢してた。 Wikiは集合知って言ってた人のことバカにしてたよね? ソースはWiki、よりも信ぴょう性にかけるLLMを集合知??????? 恥のほうじゃなくて?… | 研修で同僚がLLMのことを人工知能って言い張ってる時に「AIは集合知だから!」って何度も言っててめっちゃ笑うのを我慢してた。 Wikiは集合知って言ってた人のことバカにしてたよね? ソースはWiki、よりも信ぴょう性にかけるLLMを集合知??????? 恥のほうじ… | ↗Not briefed |
bluesky bluesky-ai | 2mo | So often I encounter people who are astounded by LLM output they interpret as especially smart or human. I’m like, yeah… it trained on stac… | So often I encounter people who are astounded by LLM output they interpret as especially smart or human. I’m like, yeah… it trained on… | ↗Not briefed |
bluesky bluesky-ai | 2mo | As AI firms push for greater controls on the industry, Al Jazeera unpacks the value of the main companies in the sector. | As AI firms push for greater controls on the industry, Al Jazeera unpacks the value of the main companies in the sector. | ↗Not briefed |
arxiv arxiv-ai | 2mo | Constraint-Safe Graph-Context Scoring for Stable Point-Feature Labels Under Text-Width and Accessibility-Inspired Profiles | Point-feature label placement on interactive maps must reconcile geometric validity, display yield, local placement utility, and stabi… | ↗Not briefed |
arxiv arxiv-ai | 2mo | Improving Cross-embodiment Transfer in Latent Action Models with Action-Similarity Supervision | As generalist robot policies gain vision and language from web-scale pretraining, demonstrations remain costly to collect and tied to… | ↗Not briefed |
arxiv arxiv-ai | 2mo | Trust, but Validate the Instrument: Auditing AI-Generated RTL Verification Plans on Authored Security-Regression Proxies | AI-generated RTL verification plans can satisfy a provider schema yet fail at the boundary to trusted execution. We present SecTB-RTL,… | ↗Not briefed |
arxiv arxiv-ai | 2mo | A Dual-Process Perspective on Nudge Susceptibility in LLM-Based GUI Agents | LLM-based GUI agents increasingly act on behalf of users in digital environments that were designed with human users in mind. These gr… | ↗Not briefed |
bluesky bluesky-ai | 2mo | I agree, although I think there's also a component of "paying someone to know about all the stupid, irrational gotchas in English Law" whic… | I agree, although I think there's also a component of "paying someone to know about all the stupid, irrational gotchas in English Law"… | ↗Not briefed |