Difference between revisions of "AI benchmarks"

From GISAXS
Jump to: navigation, search
(Assistant/Agentic)
(Hallucination)
 
(One intermediate revision by the same user not shown)
Line 19: Line 19:
 
** 2025-04: [https://x.com/scaling01/status/1910499781601874008 add] Quasar Alpha, Optimus Alpha, Llama-4 Scout and Llama-4 Maverick
 
** 2025-04: [https://x.com/scaling01/status/1910499781601874008 add] Quasar Alpha, Optimus Alpha, Llama-4 Scout and Llama-4 Maverick
 
* [https://arxiv.org/abs/2502.01100 ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning]. Assess reasoning using puzzles of tunable complexity.
 
* [https://arxiv.org/abs/2502.01100 ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning]. Assess reasoning using puzzles of tunable complexity.
 +
* 2025-08: [https://arxiv.org/abs/2508.12790 Reinforcement Learning with Rubric Anchors] ([https://huggingface.co/inclusionAI/Rubicon-Preview model])
  
 
==Task Length==
 
==Task Length==
Line 35: Line 36:
 
* [https://www.vectara.com/ Vectara] [https://github.com/vectara/hallucination-leaderboard Hallucination Leaderboard]
 
* [https://www.vectara.com/ Vectara] [https://github.com/vectara/hallucination-leaderboard Hallucination Leaderboard]
 
* [https://github.com/lechmazur/confabulations/ LLM Confabulation (Hallucination) Leaderboard for RAG]
 
* [https://github.com/lechmazur/confabulations/ LLM Confabulation (Hallucination) Leaderboard for RAG]
 +
* 2025-09: OpenAI: [https://cdn.openai.com/pdf/d04913be-3f6f-4d2b-b283-ff432ef4aaa5/why-language-models-hallucinate.pdf Why Language Models Hallucinate]
  
 
==Software/Coding==
 
==Software/Coding==

Latest revision as of 09:11, 8 September 2025

General

Lists of Benchmarks

Analysis of Methods

Methods

Task Length

GmZHL8xWQAAtFlF.jpeg

Assess Specific Attributes

Various

Hallucination

Software/Coding

Math

Science

Visual

Conversation

Creativity

Reasoning

Assistant/Agentic

See: AI Agents: Optimization

Science

See: Science Benchmarks