Google Research Details Antigravity /teamwork Maths Results
Google reports Antigravity /teamwork Lean formalisations, including Courtade–Kumar. Most results are preprints; Navier–Stokes is not claimed as solved.
Latest
Timer-M1 Preprint Says It Ranks First on Two Forecasting Benchmarks, With Its Authors’ Own Caveats
A preprint says Timer-M1, a forecasting model, ranks first on FEV and TIME. Its authors call the TIME margin over TimesFM-3 narrow. Not peer reviewed.
Quantum Roundup: A 42-Qubit Molecule Simulation, a Chip That Holds Photons, and Oratomic’s $775 Million
Three labelled quantum items: a 42-qubit molecule simulation (with a classical caveat), a chip photon memory over 1 microsecond, and Oratomic's $775M.
IonQ Says It Linked a Trapped-Ion Qubit to a Solid-State Memory at More Than 1,000 Entangled Pairs a Second
IonQ says it linked a trapped-ion qubit to a solid-state memory at over 1,000 entangled pairs a second. A preprint backs it. What it does and does not show.
ReSI Preprint Says Automated Retraining Cut Attack Success to 31%
A preprint says its ReSI method cut mean X-Teaming attack success across four models from 86.01% to 31.45%. The authors' own tests; not peer reviewed.
Do AI Agents Overstep or Fib? Two New Ways of Checking
Two new attempts to measure whether AI agents behave: Arena's Alignment Index preview, which scores 27 models on real user sessions (Arena's own claims, published the day it raised $200 million), and FAR.AI's "Caught in the Act" preprint on probes that read a model's internals to spot deception. What each found, and the limits.
Minimal Coding Agent Matches Elaborate AI Harnesses, Preprint Finds
A preprint from EPFL researchers, listed on Apple's research site, finds a single minimal coding-agent session matched or beat elaborate machine-learning agent systems in its tests. Pasqal releases skills that let coding agents prepare quantum experiments, and Microsoft makes 30 Dynamics 365 CRM skills generally available, with Autopilot still in private preview.
Claude Agents Help Build a UV Sky Map as Studies Question Agent Scores
Anthropic says a team of Claude Science agents helped build the first complete UV map of the sky, about a third of it predicted rather than observed. A preprint finds agent benchmark scores can move because of the grader, not the agent, and Exa previews a search benchmark it designed and runs itself. What each one shows, and what it doesn't.
Robotics Roundup: How Robot World Models Scale, a Beijing Funding Round and a Shop-Floor Humanoid
Three robotics items from 8 October 2026: a FAIR at Meta and Mila preprint on how robot "world models" scale, a reported Series A for Beijing's Livsyn Robotics, and CP AI Robotics' C1 store humanoid, which the company plans to roll out in Japanese convenience stores "as early as 2027". What each shows, and what it doesn't.
AI Research Tools: TasteVal Measures “Research Taste”, Karotte Targets Reward Hacking and Zephon Fixes Data Order
A new benchmark, TasteVal, says the best AI model now reaches expert-level results on eight AI research tasks with 2.3 times less compute, though its authors list reasons that may overstate progress. Preference Model has open-sourced Karotte, a framework for training environments that are harder to game, and DatologyAI has released Zephon, a data loader that keeps experiments repeatable when the GPU count changes. What each claims, and who measured it.
AI Trails People on Visual Reasoning Test; Base Models Reason on Cue
Scale AI and Elorian's Humanity's Sixth Sense benchmark finds people score 93.1% on everyday visual questions while the best model, GPT-6 Astra, scores 53.6%, mostly because models misread the picture. A separate preprint, not yet peer reviewed, says base models can reason almost as well as RL-trained ones when their answer starts with the right word.
Timeline
Every story in preprint, grouped by month.
November 2026
- 14 Nov 2026: Google Research Details Antigravity /teamwork Maths ResultsGoogle reports Antigravity /teamwork Lean formalisations, including Courtade–Kumar. Most results are preprints; Navier–Stokes is not claimed as solved.
October 2026
- 27 Oct 2026: Pasqal Says 256-Qubit Processor Found Unexpected Magnet BehaviourPasqal says its 256-qubit neutral-atom processor caught a quantum magnet settling far slower than predicted. It makes no quantum advantage claim.
- 24 Oct 2026: Mila and NRCan AI Maps Wildfire Hazard Hundreds of Times FasterMila and Natural Resources Canada say an AI copy of the BurnP3+ simulator maps wildfire hazard in seconds. The preprint is not yet peer…
- 21 Oct 2026: OpenAI Says GPT-Rosalind Agents Mined a Brain Atlas in 32 MinutesOpenAI, PerturbAI and Scripps say 1,946 GPT-Rosalind agents turned a mouse-brain CRISPR atlas into research leads in 32 minutes. Lab tests come next.
- 15 Oct 2026: Google Research Makes Deep Review Public, an AI Agent That Checks Technical DocumentsGoogle Research made Deep Review public, an AI agent that checks technical documents. Users get two free reviews a week.
- 10 Oct 2026: Timer-M1 Preprint Says It Ranks First on Two Forecasting Benchmarks, With Its Authors' Own CaveatsA preprint says Timer-M1, a forecasting model, ranks first on FEV and TIME. Its authors call the TIME margin over TimesFM-3 narrow. Not peer…
- 10 Oct 2026: Quantum Roundup: A 42-Qubit Molecule Simulation, a Chip That Holds Photons, and Oratomic's $775 MillionThree labelled quantum items: a 42-qubit molecule simulation (with a classical caveat), a chip photon memory over 1 microsecond, and Oratomic's $775M.
- 10 Oct 2026: IonQ Says It Linked a Trapped-Ion Qubit to a Solid-State Memory at More Than 1,000 Entangled Pairs a SecondIonQ says it linked a trapped-ion qubit to a solid-state memory at over 1,000 entangled pairs a second. A preprint backs it. What it…
- 10 Oct 2026: ReSI Preprint Says Automated Retraining Cut Attack Success to 31%A preprint says its ReSI method cut mean X-Teaming attack success across four models from 86.01% to 31.45%. The authors' own tests; not peer…
- 10 Oct 2026: Do AI Agents Overstep or Fib? Two New Ways of CheckingTwo new attempts to measure whether AI agents behave: Arena's Alignment Index preview, which scores 27 models on real user sessions (Arena's own claims,…
- 9 Oct 2026: Minimal Coding Agent Matches Elaborate AI Harnesses, Preprint FindsA preprint from EPFL researchers, listed on Apple's research site, finds a single minimal coding-agent session matched or beat elaborate machine-learning agent systems in…
- 9 Oct 2026: Claude Agents Help Build a UV Sky Map as Studies Question Agent ScoresAnthropic says a team of Claude Science agents helped build the first complete UV map of the sky, about a third of it predicted…
Load more Showing 12 of 16
Related stories
Latest from topics connected to preprint. See all with connected topics
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.
