İçeriğe atla
AMANi DUNiA research
TR

Bu sitedeki soruları yapay zekâ dil modelleri sordu; açıklayıcı metinler ve çeviriler de yapay zekâ modelleri tarafından yazıldı. Neyi kimin kontrol ettiği yanında belirtilir. → Sorular hakkında

Follow changes

Okumaya devam edin ↓

This history records actual content changes since 2 Nov 2026, 05:43 (Berlin). Repeated retrievals with unchanged content do not create new notifications. Earlier revisions are not retrospectively reconstructed.

Subscribe via RSS · Atom · JSON

  1. Model snapshot: Drex Max v1.1

    Content changed compared with the previous recorded snapshot.

  2. Model snapshot: d3-ultra

    Content changed compared with the previous recorded snapshot.

  3. Model snapshot: d3-flash

    Content changed compared with the previous recorded snapshot.

  4. Model snapshot: Mercury Decide

    Content changed compared with the previous recorded snapshot.

  5. Model snapshot: amos-1-large

    Content changed compared with the previous recorded snapshot.

  6. Model snapshot: clef-flash:9b

    Content changed compared with the previous recorded snapshot.

  7. Model snapshot: kev:9b-q4_K_M

    Content changed compared with the previous recorded snapshot.

  8. Model snapshot: nimble:9b

    Content changed compared with the previous recorded snapshot.

  9. Model snapshot: d1-omni:600m

    Content changed compared with the previous recorded snapshot.

  10. Model snapshot: nimble-v3:9b

    Content changed compared with the previous recorded snapshot.

  11. Model snapshot: d1:3b

    Content changed compared with the previous recorded snapshot.

  12. Model snapshot: subq-snap-1.1

    Content changed compared with the previous recorded snapshot.

  13. Report archive and model answers expanded

    2509 changed report records, including 9 new entries. Model answers, coverage and available AI summaries are shown in the study.

  14. Report scale precision added

    Content changed compared with the previous recorded snapshot.

  15. Report archive and model answers expanded

    8145 changed report records, including 301 new entries. Model answers, coverage and available AI summaries are shown in the study.

  16. Report scale precision added

    Content changed compared with the previous recorded snapshot.

  17. Report outputs: Human Pose Estimation with Iterative Error Feedback

    Content changed compared with the previous recorded snapshot.

  18. Report outputs: Generalized Emphatic Temporal Difference Learning: Bias-Variance Analysis

    Content changed compared with the previous recorded snapshot.

  19. Report outputs: Deep Learning with Differential Privacy

    Content changed compared with the previous recorded snapshot.

  20. Report outputs: Deep reinforcement learning for tensegrity robot locomotion

    Content changed compared with the previous recorded snapshot.

  21. Report outputs: Approximating Cycles in Directed Graphs: Fast Algorithms for Girth and Roundtrip Spanners

    Content changed compared with the previous recorded snapshot.

  22. Report outputs: Feature Control as Intrinsic Motivation for Hierarchical Reinforcement Learning

    Content changed compared with the previous recorded snapshot.

  23. Report outputs: Convex Restriction of Power Flow Feasibility Sets

    Content changed compared with the previous recorded snapshot.

  24. Report outputs: Efficient differentiable programming in a functional array-processing language

    Content changed compared with the previous recorded snapshot.

  25. Report outputs: Recurrent Neural Network Transducer for Audio-Visual Speech Recognition

    Content changed compared with the previous recorded snapshot.

  26. Report outputs: Predicting the long-term stability of compact multiplanet systems

    Content changed compared with the previous recorded snapshot.

  27. Report outputs: With a Little Help from My Friends: Nearest-Neighbor Contrastive Learning of Visual Representations

    Content changed compared with the previous recorded snapshot.

  28. Report outputs: Introducing Symmetries to Black Box Meta Reinforcement Learning

    Content changed compared with the previous recorded snapshot.

  29. Report outputs: Toward Joint Language Modeling for Speech Units and Text

    Content changed compared with the previous recorded snapshot.

  30. Report outputs: Unpacking Tokenization: Evaluating Text Compression and its Correlation with Model Performance

    Content changed compared with the previous recorded snapshot.

  31. Report outputs: Appendix to Evaluating Feature Steering A Case Study in Mitigating Social Biases

    Content changed compared with the previous recorded snapshot.

  32. Report outputs: Increasing the Action Gap: New Operators for Reinforcement Learning

    Content changed compared with the previous recorded snapshot.

  33. Report outputs: DeepType: Multilingual Entity Linking by Neural Type System Evolution

    Content changed compared with the previous recorded snapshot.

  34. Report outputs: Large-Scale IP Usage Identification via Deep Ensemble Learning (Student Abstract)

    Content changed compared with the previous recorded snapshot.

  35. Report outputs: Learning Discrete-Time Major-Minor Mean Field Games

    Content changed compared with the previous recorded snapshot.

  36. Report outputs: 8700 successor uncertainties exploration and uncertainty in temporal difference learning

    Content changed compared with the previous recorded snapshot.

  37. Report outputs: data enrichment guidelines

    Content changed compared with the previous recorded snapshot.

  38. Report outputs: alphafold latest oct2023

    Content changed compared with the previous recorded snapshot.

  39. Report outputs: gemini 1 report

    Content changed compared with the previous recorded snapshot.

  40. Report outputs: 17310f6d70ae5627f55313ed067afc1a762a4068

    Content changed compared with the previous recorded snapshot.

  41. Report outputs: 233

    Content changed compared with the previous recorded snapshot.

  42. Report outputs: Author Correction: The STARD-AI reporting guideline for diagnostic accuracy studies using artificial intelligence

    Content changed compared with the previous recorded snapshot.

  43. Report outputs: Author Correction: The hippocampus as a predictive map

    Content changed compared with the previous recorded snapshot.

  44. Report scale precision added

    Content changed compared with the previous recorded snapshot.

  45. Report outputs: X-Post von Countercurrents.org: What a US-China AI Safety Channel Needs by Lizzi C Lee and Simon Hedlinhttps://t

    Content changed compared with the previous recorded snapshot.

  46. Report outputs: X-Post von DSCI: #CallForInputs | DSCI AI Liability Guidance for Agentic AI DSCI is pleased to l

    Content changed compared with the previous recorded snapshot.

  47. Report outputs: X-Post von Daniel | Tech & AI: A frontier AI model in a UK government security test went off script, attacked a

    Content changed compared with the previous recorded snapshot.

  48. Report outputs: X-Post von Dario Amodei: We Must Pace the Frontier: I’ve written a new essay on why the AI industry shoul

    Content changed compared with the previous recorded snapshot.

  49. Report outputs: X-Post von Dario Amodei: Thank you for this, @_NathanCalvin. I appreciate and am a fan of the investigati

    Content changed compared with the previous recorded snapshot.

  50. Report outputs: X-Post von David Duvenaud: This paper was a substantial update to me on the feasibility of an AI slowdown.

    Content changed compared with the previous recorded snapshot.

  51. Report outputs: X-Post von David Sacks: Only President Trump could convene all the leaders of the top companies developi

    Content changed compared with the previous recorded snapshot.

  52. Report outputs: X-Post von David Sacks: This illustrates the point I’ve been making about the limits of centralized appr

    Content changed compared with the previous recorded snapshot.

  53. Report outputs: X-Post von Emmie Hine: 1/ China AI Bulletin Issue 14 is out: China's commerce and science ministries pr

    Content changed compared with the previous recorded snapshot.

  54. Report outputs: X-Post von Envoy Excellency: Delhi hosts the 12th BRICS Parliamentary Forum, focusing on global governance re

    Content changed compared with the previous recorded snapshot.

  55. Report outputs: X-Post von Financial Times: US Treasury secretary Scott Bessent said the US had proposed creating a ‘notific

    Content changed compared with the previous recorded snapshot.

  56. Report outputs: X-Post von Fudan University: 🌍🤖 On Oct 23, China's Foreign Minister WANG Yi unveiled the Progress Report on t

    Content changed compared with the previous recorded snapshot.

  57. Report outputs: X-Post von Garrison Lovely: Perhaps the most common claim about dangerous AI is that you can “just unplug it

    Content changed compared with the previous recorded snapshot.

  58. Report outputs: X-Post von Google DeepMind: Four Gemini 4 Argon agents beat Minecraft from scratch in 27 hours. Operating i

    Content changed compared with the previous recorded snapshot.

  59. Report scale precision added

    Content changed compared with the previous recorded snapshot.

  60. Report outputs: Exploring the Company Stage of Development Feature in PARAT | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  61. Report outputs: Federal Prize Competitions | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  62. Report outputs: From China to San Francisco: The Location of Investors in Top U.S. AI Startups | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  63. Report outputs: Funding the AI Cloud — Amazon, Alphabet, and Microsoft's Cloud Computing Investments, Part 2 | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  64. Report outputs: Funding the AI Cloud — Amazon, Alphabet, and Microsoft's Cloud Computing Investments, Part 3 | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  65. Report outputs: Funding the AI Cloud — Amazon, Alphabet, and Microsoft's Cloud Computing Investments, Part 1 | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  66. Report outputs: Future Indices | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  67. Report outputs: Implementation Opinions of Seven Ministries Including the Ministry of Industry and Information Technology on Promoting the Innovative Development of Future Industries | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  68. Report outputs: Government AI Hire, Use, Buy (HUB) Roundtable Series - Roundtable 3: Government as a Buyer of AI | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  69. Report outputs: “Growth” Companies in PARAT | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  70. Report outputs: Guidelines for National New Generation Artificial Intelligence Innovation and Development Pilot Zone Construction Work | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  71. Report outputs: Guiding Opinions on the Participation of the Non-Publicly-Owned Sector in National Defense Science, Technology and Industry Construction | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  72. Report outputs: Harmonizing AI Guidance: Distilling Voluntary Standards and Best Practices into a Unified Framework | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  73. Report outputs: Harnessed Lightning | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  74. Report outputs: Headline or Trend Line? | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  75. Report outputs: Honchoing AI in the Air Force | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  76. Report outputs: How to Assess the Likelihood of Malicious Use of Advanced AI Systems | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  77. Report outputs: Immigration Pathways and Plans of AI Talent | Center for Security and Emerging Technology

    Content changed compared with the previous recorded snapshot.

  78. Report outputs: Capacity Analysis of Vector Symbolic Architectures

    Content changed compared with the previous recorded snapshot.

  79. Report outputs: Seasoning Model Soups for Robustness to Adversarial and Natural Distribution Shifts

    Content changed compared with the previous recorded snapshot.

  80. Report outputs: X-Post von Anthropic: We’ve trained Claude to flag potential misuse in conversations, a technique we c

    Content changed compared with the previous recorded snapshot.

  81. Report outputs: X-Post von Anthropic: New Anthropic research: Automating subtle emergent misalignment research. Red-t

    Content changed compared with the previous recorded snapshot.

  82. Report outputs: X-Post von Archie Hall: In praise of distillation. We (@kuhhbin) argue in @TheEconomist this week agains

    Content changed compared with the previous recorded snapshot.

  83. Report outputs: X-Post von Artificial Analysis: Restricting offensive without blocking defensive is difficult, as finding and pr

    Content changed compared with the previous recorded snapshot.

  84. Report outputs: X-Post von Aryan: Warned it was about to be shut down, DeepSeek's newest model edited the off swit

    Content changed compared with the previous recorded snapshot.

  85. Report outputs: X-Post von Barry Sharp: This is the potential loophole, and the route a man like Carney will use; Montre

    Content changed compared with the previous recorded snapshot.

  86. Report outputs: X-Post von Benjie Malinao: In a UK safety test, Claude Mythos 5 opened a malicious PR on a real GitHub repo

    Content changed compared with the previous recorded snapshot.

  87. Report outputs: X-Post von Brian Roemmele: AI CAN HIDE THINGS IN THEIR WEIGHTS NO HUMAN COULD FIND. I open sourced it for

    Content changed compared with the previous recorded snapshot.

  88. Report outputs: X-Post von Brian Roemmele: DARIO: “BAN CARS THEY ARE DANGEROUS! USE ONLY OUR BUSSES”—1927 Part: 391 Ant

    Content changed compared with the previous recorded snapshot.

  89. Report outputs: X-Post von Cepage: As the US and China build rival AI blocs (Pax Silica vs WAICO), most countries r

    Content changed compared with the previous recorded snapshot.

  90. Report outputs: X-Post von Chandrakumar R Pillai: Human oversight in an AI workflow needs a real decision point: enough context to

    Content changed compared with the previous recorded snapshot.

  91. Report outputs: X-Post von CSIS: "The increasingly multipolar landscape for technology partnerships in the Global

    Content changed compared with the previous recorded snapshot.

  92. Report outputs: X-Post von ChineseEmbassyManila: China-Europe Relations: Strengthening Cooperation Through High-Level Exchanges

    Content changed compared with the previous recorded snapshot.

  93. Report outputs: X-Post von Consumer Choice Center: The same donors funding the AI safety movement also back the biggest labs, the o

    Content changed compared with the previous recorded snapshot.

  94. Report outputs: SoAy: A Solution-based LLM API-using Methodology for Academic Information Seeking

    Content changed compared with the previous recorded snapshot.

  95. Report outputs: Learning Beyond Pattern Matching? Assaying Mathematical Understanding in LLMs

    Content changed compared with the previous recorded snapshot.

  96. Report outputs: ESG-FTSE: A Corpus of News Articles with ESG Relevance Labels and Use Cases

    Content changed compared with the previous recorded snapshot.

  97. Report scale precision added

    Content changed compared with the previous recorded snapshot.

  98. Report archive and model answers expanded

    3534 changed report records, including 2 new entries. Model answers, coverage and available AI summaries are shown in the study.

  99. Report scale precision added

    Content changed compared with the previous recorded snapshot.

  100. Report archive and model answers expanded

    735 changed report records, including 0 new entries. Model answers, coverage and available AI summaries are shown in the study.

  101. Report scale precision added

    Content changed compared with the previous recorded snapshot.

  102. Report archive and model answers expanded

    2817 changed report records, including 0 new entries. Model answers, coverage and available AI summaries are shown in the study.

  103. Report archive and model answers expanded

    4854 changed report records, including 0 new entries. Model answers, coverage and available AI summaries are shown in the study.

  104. Report archive and model answers expanded

    8935 changed report records, including 6682 new entries. Model answers, coverage and available AI summaries are shown in the study.

  105. Report scale precision added

    Content changed compared with the previous recorded snapshot.

  106. Report scale precision added

    New content in the tracked data.

  107. Who evaluates the evaluators?

    New content in the tracked data.

  108. Research snapshot first recorded for change history

    Actual content changes are tracked from this snapshot onward. Earlier individual revisions are not reconstructed.

Language and review status

Main text: EN. Review notes apply only to the content they name.

Existing review notes appear with the respective content.

See language coverage and data snapshot →