Articles

8 articles
Tag: reasoning ✕
MCircKE: Mechanistic Circuit-Based Knowledge Editing Bridges the Reasoning Gap in LLMs ML
2026-04-08T05:20:34.202Z · Src: 2026-04-08T00:00:00.000Z
llmknowledge editingmechanistic interpre
Why Parallel Sampling Beats Sequential Sampling in AI Reasoning Models AI
2026-04-08T05:14:57.381Z · Src: 2026-04-08T00:00:00.000Z
aireasoningllm
Cog-DRIFT: Teaching LLMs to Learn from Problems They Can't Yet Solve Through Task Reformulation AI
2026-04-07T19:53:03.673Z · Src: 2026-04-07T00:00:00.000Z
llmreinforcement learnirlvr
Bidirectional Entropy Modulation: Rethinking Exploration in Reinforcement Learning for LLM Reasoning AI
2026-04-07T17:16:22.212Z · Src: 2026-04-07T00:00:00.000Z
reinforcement learnillmreasoning
The Neurosymbolic AI Convergence: Combining Neural Networks With Logical Reasoning for Next-Generation AI AI
2026-04-04T19:26:02.216Z · Src: 2026-04-04T00:00:00.000Z
neurosymbolic aiaireasoning
Anthropic Preparing New Models Codenamed Mythos and Capybara AI
2026-03-27T08:22:47.694Z · Src: 2026-03-27T00:00:00.000Z
anthropicclaudemythos
Symbolica's Agentica SDK Achieves 36% on ARC-AGI-3 Day 1, Crushing CoT Baselines at 1/9th Cost AI
2026-03-27T02:16:32.996Z · Src: 2026-03-27T00:00:00.000Z
aiarc agi 3symbolica
ARC-AGI-3: The Next Frontier of Artificial General Intelligence Benchmarking AI
2026-03-26T00:19:50.000Z · Src: 2026-03-26T00:00:00.000Z
ARC Prize announces ARC-AGI-3, the third iteration of its influential AGI benchmark testing novel visual reasoning. The new version features more complex puzzles and a higher difficulty ceiling for ev
agiarcbenchmark