AINews Portal

News

50 items (filtered)

The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.

SourceAgeTitleSummaryStatus
arxiv
arxiv-ai
11dPyrite: Building and Benchmarking Protein-Ligand Pose Ranking AgentsPredicting how a ligand binds to its target is central to small-molecule drug discovery. Modern cofolding models can generate diverse…
↗Not briefed
hn
hn-ai
11dAnthropic Has No Moat—
↗Not briefed
bluesky
bluesky-ai
11dlike once again not a coder but isn’t finding ways to solve a problem or circumvent an issue kind of like… the primary joy of doing it?? is…like once again not a coder but isn’t finding ways to solve a problem or circumvent an issue kind of like… the primary joy of doing it…
↗Not briefed
hn
hn-ai
11dFive Self-Contained CUDA Optimizations for 3D Gaussian Splatting—
↗Not briefed
bluesky
bluesky-ai
11dI'm forever grateful that, having followed the development of machine learning technology for over a decade out of academic interest, the m…I'm forever grateful that, having followed the development of machine learning technology for over a decade out of academic interest,…
↗Not briefed
arxiv
arxiv-ai
11dWhen Does the Real Wiring Win? Testing connectome structure-function claims against unmeasured physiologyConnectome-constrained models are often used to argue that a measured wiring diagram outperforms shuffled or rewired wiring, and hence…
↗Not briefed
bluesky
bluesky-ai
11dand no an LLM isn’t driving the villagers’ AI, it’s my own variant of GOAP that i think i’m going to call Goose Oriented Action Planningand no an LLM isn’t driving the villagers’ AI, it’s my own variant of GOAP that i think i’m going to call Goose Oriented Action Planni…
↗Not briefed
hn
hn-ai
11dTool for managing RAM by tracking RAM usage by Claude sessions and Chrome tabs—
↗Not briefed
arxiv
arxiv-ai
11dContinual-ARC: Measuring Skill Acquisition over a Working LifetimeAgent harnesses have emerged as a promising mechanism for continual learning, alongside test-time parametric learning. However, the cu…
↗Not briefed
bluesky
bluesky-ai
11dIt seems extremely clear to me that the response to BIG WALK using LLM tools for bugfixes will lead to studios never disclosing their LLM u…It seems extremely clear to me that the response to BIG WALK using LLM tools for bugfixes will lead to studios never disclosing their…
↗Not briefed
bluesky
bluesky-ai
11dI feel like if you disqualify Big Walk from GOTY discussion because they used an LLM to solve an otherwise intractable technical problem in…🔧 ConsensusI feel like if you disqualify Big Walk from GOTY discussion because they used an LLM to solve an otherwise intractable technical probl…
↗Not briefed
bluesky
bluesky-ai
11dlike as much as i would like to be like that nerd ass game y’all niggas run used ai #SAD its just really. nothing its such a none issue vs …like as much as i would like to be like that nerd ass game y’all niggas run used ai #SAD its just really. nothing its such a none issu…
↗Not briefed
hn
hn-ai
11dHow Curative is building a national provider network with agents on Claude—
↗Not briefed
arxiv
arxiv-ai
11dEmergent Biological Capabilities in a Foundation Model for Molecular InteractionsFoundation models trained on large unlabeled corpora develop emergent capabilities, tasks that go beyond the training objective. We ex…
↗Not briefed
bluesky
bluesky-ai
11dI know it's unavoidable - Win 10, NV drivers, VS2019, etc. no doubt have had some LLM involved, but I don't pay for that LOL and I get to u…I know it's unavoidable - Win 10, NV drivers, VS2019, etc. no doubt have had some LLM involved, but I don't pay for that LOL and I get…
↗Not briefed
bluesky
bluesky-ai
11dI'm gonna be real even if the reason is a bit stupid, after Reading The Article I don't really see it as meaningful llm usageI'm gonna be real even if the reason is a bit stupid, after Reading The Article I don't really see it as meaningful llm usage
↗Not briefed
bluesky
bluesky-ai
11di have used an LLM before, i asked it two questions about forth in 2024. which is cool for me because 1) it didn't answer my questions and …i have used an LLM before, i asked it two questions about forth in 2024. which is cool for me because 1) it didn't answer my questions…
↗Not briefed
rss
techmeme
11dSources: Citigroup, Goldman Sachs, and others in a 21-bank consortium committed around $270M to fund five years of a stablecoin venture announced in September (Bloomberg)Bloomberg: Sources: Citigroup, Goldman Sachs, and others in a 21-bank consortium committed around $270M to fund five years of a stable…
↗Not briefed
arxiv
arxiv-ai
11dComparative Analysis of Agent Evaluation Frameworks: Stability, Detection, and Discovery in Enterprise Analytics AgentsDeploying LLM-based analytics agents in enterprise settings requires evaluation frameworks that can reliably detect failures across co…
↗Not briefed
rss
techcrunch-ai
11dOpenAI rolls out new business profiles, making ChatGPT just a little bit more like Google🔧 ChatGPTOpenAI is looking to make ChatGPT a more robust consumer discovery engine, and a host of new features seek to empower businesses to he…
↗Not briefed
arxiv
arxiv-ai
11dReTrace: Distilling Tool-Using Agents into Planner-Free WorkflowsBusiness process automation runs the same Standard Operating Procedure (SOP) again and again on new inputs. Large Language Model (LLM)…
↗Not briefed
arxiv
arxiv-ai
11dTimeSeek: Multi-Task and Multi-Domain Pretraining for Generalized Time Series AnalyticsTime series analytics has been widely applied across various domains. While existing time series foundation models (TSFMs) have demons…
↗Not briefed
hn
hn-ai
11dPutting a Wired Device on WiFi—
↗Not briefed
bluesky
bluesky-ai
11dthis is not math, this is programming, and as little as I trust OpenAI that their tools discovered new mathematics, "AI solved previously u…this is not math, this is programming, and as little as I trust OpenAI that their tools discovered new mathematics, "AI solved previou…
↗Not briefed
bluesky
bluesky-ai
11dthe google search LLM like genuinely gives me what i need 95% of the time. if it doesn't, i just ask a follow up question. its faster than …🔧 ChatGPTthe google search LLM like genuinely gives me what i need 95% of the time. if it doesn't, i just ask a follow up question. its faster…
↗Not briefed
bluesky
bluesky-ai
11dim not necessarily disagreeing that google went downhill, but i will say that ive been using google more now than i use claude/chatgpt to d…🔧 ChatGPTim not necessarily disagreeing that google went downhill, but i will say that ive been using google more now than i use claude/chatgpt…
↗Not briefed
arxiv
arxiv-ai
11dRe-Search: Dissecting LLM Ideation Through a Two-Player Questioning GameResearch ideation is iterative: researchers refine their ideas step by step based on feedback. Yet existing evaluations obscure this s…
↗Not briefed
bluesky
bluesky-ai
11dAnd so now you get LLM output that is full top to bottom with turns of phrase that hit really nicely as a highlight of otherwise sparse and…And so now you get LLM output that is full top to bottom with turns of phrase that hit really nicely as a highlight of otherwise spars…
↗Not briefed
bluesky
bluesky-ai
11dI don't personally think there's any ethical use case for LLM, and even debugging can lead to cases where your programmers can end up strug…🔧 ReplicateI don't personally think there's any ethical use case for LLM, and even debugging can lead to cases where your programmers can end up…
↗Not briefed
bluesky
bluesky-ai
11dthe reaction to the big walk thing has me in a sour mood. it feels like its being judged hypervigilantly because collectively we want to be…the reaction to the big walk thing has me in a sour mood. it feels like its being judged hypervigilantly because collectively we want…
↗Not briefed
arxiv
arxiv-ai
11dMechanistic Interpretability Needs a Logic of InferenceNeural networks are increasingly studied not merely as predictive systems, but as complex systems whose internal computations are them…
↗Not briefed
bluesky
bluesky-ai
11dI'm sure many (most) software engineers are in the same boat I am, where we're *required* to use LLM agents, even in places where it does n…I'm sure many (most) software engineers are in the same boat I am, where we're *required* to use LLM agents, even in places where it d…
↗Not briefed
bluesky
bluesky-ai
11dI'm now in the position of "will I willingly play a game with LLM in it" and the answer is no, as it was when I first found out about a Po…I'm now in the position of "will I willingly play a game with LLM in it" and the answer is no, as it was when I first found out about…
↗Not briefed
bluesky
bluesky-ai
11dSide note: I am also a bit more sympathetic to people bug testing their games with AI if it connects people over the internet. All it take…Side note: I am also a bit more sympathetic to people bug testing their games with AI if it connects people over the internet. All it…
↗Not briefed
hn
hn-ai
11dMore Ways to Work with Claude: Sheets, Whiteboards, and Interactive Connectors—
↗Not briefed
bluesky
bluesky-ai
11dThank you UKRAINE‼️💙♥️🇺🇦🇺🇸 Timing is everything -October surprise Ukrainian Style😈☠️😎 Bot traffic from ‘Russia’s Google’ nearly wipe…Thank you UKRAINE‼️💙♥️🇺🇦🇺🇸 Timing is everything -October surprise Ukrainian Style😈☠️😎 Bot traffic from ‘Russia’s Google’ nearly…
↗Not briefed
bluesky
bluesky-ai
11dAnthropic’s seven founders plan to cash out about $250 million apiece in the IPO—leaving them with stakes worth around $18 billion each. on…Anthropic’s seven founders plan to cash out about $250 million apiece in the IPO—leaving them with stakes worth around $18 billion eac…
↗Not briefed
bluesky
bluesky-ai
11dlike i don't want a government telling people what software they can make or use. but i do want them telling anthropic, google, openai etc…like i don't want a government telling people what software they can make or use. but i do want them telling anthropic, google, opena…
↗Not briefed
rss
the-decoder
11dNvidia's model GR00T 2 watches humans on video, then imagines itself as the robotNvidia's new AI model GR00T 2 is designed to teach robots new tasks without laborious teleoperation. Instead, the system transfers hum…
↗Not briefed
arxiv
arxiv-ai
11dPterodactyl: Pushing the Limits of Efficient Automated Translation EvaluationIn this paper, we present 1020 Group's submission to the WMT26 shared task on automated quality evaluation. Our system, Pterodactyl, w…
↗Not briefed
hn
hn-ai
11dHow do you keep up quality when you vibe-code?I'm having a really hard time working with claude. In the past when I wrote all the code manually, it was always carefully review…
↗Not briefed
hn
hn-ai
11dSpec-Driven Development for Codex—
↗Not briefed
bluesky
bluesky-ai
11dAnthropic brings Claude to Google Meet to tackle tasks live reut.rs/4j9DvaAAnthropic brings Claude to Google Meet to tackle tasks live reut.rs/4j9DvaA
↗Not briefed
arxiv
arxiv-ai
11dFERN: Cross-Benchmark Table Retrieval for Text-to-SQL with Frozen Retrievers and Learned RankingText-to-SQL systems over large pooled catalogs must first retrieve the tables a question needs. We compare against the published table…
↗Not briefed
bluesky
bluesky-ai
11d‘The tools allow for Claude to join a meeting, take notes, lay out ideas on a shared whiteboard, edit spreadsheets alongside employees, or …‘The tools allow for Claude to join a meeting, take notes, lay out ideas on a shared whiteboard, edit spreadsheets alongside employees…
↗Not briefed
bluesky
bluesky-ai
11dAI disclosure policies are becoming tricky to navigate not only due to AI's increasing ubiquity (e.g., if I use Wolfram Alpha to reality-ch…AI disclosure policies are becoming tricky to navigate not only due to AI's increasing ubiquity (e.g., if I use Wolfram Alpha to reali…
↗Not briefed
bluesky
bluesky-ai
11dHow cute, Anthropic is committing 0.0036% of its IPO value to food banks and Toys for Tots. Surely this will rescue its reputation! www.sf…How cute, Anthropic is committing 0.0036% of its IPO value to food banks and Toys for Tots. Surely this will rescue its reputation! w…
↗Not briefed
hn
hn-ai
11dTransformers for Mobile User Acquisition—
↗Not briefed
rss
ai-snake-oil
11dThe missing quadrant in responses to AIAnnouncing STABL: Societal Transformation and AI Bottlenecks Lab
↗Not briefed
bluesky
bluesky-ai
11dfunny thing this is that they made a minimizing cover-your-ass statement when forced to. if their LLM use was as minimal as they claim, it …funny thing this is that they made a minimizing cover-your-ass statement when forced to. if their LLM use was as minimal as they claim…
↗Not briefed