• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
Science
Announcement

AI tools for checking math research and visualizing NASA Mars data

OSVerify's developers say it found five major and 13 minor issues in 11 recent OpenAI math preprints.

Sundar PichaiSP
Fan NieFN
2 Sources, 8h ago, first seen 8h ago

TLDR

OSVerify's developers say their framework found five major and 13 minor issues in 11 recent OpenAI math preprints, all verified by human experts. The framework proposes potential errors, then checks whether they are real and substantial. Separately, Sundar Pichai says Gemini 4 Argon turned 80,000-plus NASA Mars weather logs and Ingenuity Helicopter telemetry files into visual slides about atmospheric flight mechanics in minutes.

Combined views

48.5K

2 Sources, first seen 8h ago

434 likes39 comments108 saves47 reposts

Combined views

48.5K

2 Sources, first seen 8h ago

434 likes39 comments108 saves47 reposts

Sentiment

Positive76.9%23.1%Negative

Summary

Many accounts praised Gemini for turning 80,000+ raw NASA Mars logs into clean flight-mechanics outputs thanks to its long-context reasoning, while some questioned limited Pro availability and feature constraints.

Based on 21 sentiment-bearing replies from 20 accounts across 2 conversations.

Featured Source

Sentiment

Positive76.9%23.1%Negative

Summary

Many accounts praised Gemini for turning 80,000+ raw NASA Mars logs into clean flight-mechanics outputs thanks to its long-context reasoning, while some questioned limited Pro availability and feature constraints.

Based on 21 sentiment-bearing replies from 20 accounts across 2 conversations.

2 Sources

Sundar Pichai@sundarpichaiAnyone dealing with massive data dumps knows what a headache it can be. We gave Gemini 4 Argon 80,000+ raw NASA Mars weather logs and Ingenuity Helicopter telemetry files, it distills the atmospheric flight mechanics into a deck, turning massive, messy scientific archives (including a mix of CSVs, images, pdfs, xml logs, videos) into clear visual Google Slides in minutes.8h
Fan Nie@FanNie1208AI-generated scientific output is scaling beyond available human review capacity. Scaling scientific discovery requires scaling verification. We introduce OSVerify, a framework for one-sided verification: rejecting as many invalid solutions as possible while preserving rare valid ones for expert review. We used OSVerify to referee 11 recent OpenAI math preprints, identifying 5 major and 13 minor issues, all verified by human experts. Our key insight: checking a proposed error can be easier than judging an entire solution. Instead of directly judging solution correctness, OSVerify decouples verification into error proposal (searching for potential errors) and error validation (checking whether they are real and substantial). On competition and research mathematical benchmarks, OSVerify reduces invalid solutions requiring expert review by 46% on average compared to voting under the same low-FNR constraint, or 57% with an RL-trained validator. Verification can also guide discovery. OSVerify can be integrated into existing scientific discovery workflows. Paired with an agentic math solver, it helped resolve 3 open conjectures in the Langlands program and contributed to resolving the Manin constant conjecture. All proofs were confirmed by mathematicians. Our broader goal is to provide verification-based evaluation signals for scientific discovery, enabling test-time scaling and self-improvement even when reliable automatic evaluators are unavailable. Project: https://osverify.github.io/ Paper: https://osverify.github.io/assets/papers/osverify.pdf GitHub: https://github.com/OSVerify [1/n]8h
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    2 Sources

    Sundar Pichai@sundarpichaiAnyone dealing with massive data dumps knows what a headache it can be. We gave Gemini 4 Argon 80,000+ raw NASA Mars weather logs and Ingenuity Helicopter telemetry files, it distills the atmospheric flight mechanics into a deck, turning massive, messy scientific archives (including a mix of CSVs, images, pdfs, xml logs, videos) into clear visual Google Slides in minutes.8h
    Fan Nie@FanNie1208AI-generated scientific output is scaling beyond available human review capacity. Scaling scientific discovery requires scaling verification. We introduce OSVerify, a framework for one-sided verification: rejecting as many invalid solutions as possible while preserving rare valid ones for expert review. We used OSVerify to referee 11 recent OpenAI math preprints, identifying 5 major and 13 minor issues, all verified by human experts. Our key insight: checking a proposed error can be easier than judging an entire solution. Instead of directly judging solution correctness, OSVerify decouples verification into error proposal (searching for potential errors) and error validation (checking whether they are real and substantial). On competition and research mathematical benchmarks, OSVerify reduces invalid solutions requiring expert review by 46% on average compared to voting under the same low-FNR constraint, or 57% with an RL-trained validator. Verification can also guide discovery. OSVerify can be integrated into existing scientific discovery workflows. Paired with an agentic math solver, it helped resolve 3 open conjectures in the Langlands program and contributed to resolving the Manin constant conjecture. All proofs were confirmed by mathematicians. Our broader goal is to provide verification-based evaluation signals for scientific discovery, enabling test-time scaling and self-improvement even when reliable automatic evaluators are unavailable. Project: https://osverify.github.io/ Paper: https://osverify.github.io/assets/papers/osverify.pdf GitHub: https://github.com/OSVerify [1/n]8h
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet