repoGitHubTrust 82 · PrimaryPublished 1mo agoLive · 1mo ago
Amirhosein-gh98/Gnosis
Can LLMs Predict Their Own Failures? Self-Awareness via Internal Circuits
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 49%Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMs →
- PossiblePossibly related (embedding) · 49%LLMs know when they are wrong. I made a fix relating to Anthropic's new "global workspace" paper [R] →
- PossiblePossibly related (embedding) · 46%Manufactured Confidence: How Memory Consolidation Turns Hearsay into Confident Facts →
- PossiblePossibly related (embedding) · 46%Mirage of Mastery: Memorization Tricks LLMs into Artificially Inflated Self-Knowledge - The Association for the Advancement of Artificial Intelligence →
- PossiblePossibly related (embedding) · 46%Independent Clinical Evaluation of General-Purpose LLM Responses to Signals of Suicide Risk - The Association for the Advancement of Artificial Intelligence →
- PossiblePossibly related (embedding) · 46%It only took 200 update steps to flip Qwen2.5-7B-Instruct from denying sentience to developing a robust identity of being a "sentient machine" [P] →
Implements
Covers
newsLLMs know when they are wrong. I made a fix relating to Anthropic's new "global workspace" paper [R]newsMirage of Mastery: Memorization Tricks LLMs into Artificially Inflated Self-Knowledge - The Association for the Advancement of Artificial IntelligencenewsIndependent Clinical Evaluation of General-Purpose LLM Responses to Signals of Suicide Risk - The Association for the Advancement of Artificial Intelligence
Covers (incoming)
Related across the graph
paperReinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMsnewsMirage of Mastery: Memorization Tricks LLMs into Artificially Inflated Self-Knowledge - The Association for the Advancement of Artificial IntelligencepaperManufactured Confidence: How Memory Consolidation Turns Hearsay into Confident FactsnewsIndependent Clinical Evaluation of General-Purpose LLM Responses to Signals of Suicide Risk - The Association for the Advancement of Artificial IntelligencenewsIt only took 200 update steps to flip Qwen2.5-7B-Instruct from denying sentience to developing a robust identity of being a "sentient machine" [P]newsLLMs know when they are wrong. I made a fix relating to Anthropic's new "global workspace" paper [R]