framework
active
framework:axbenchAxBench
Framework from Wu et al. for automated contrastive pair generation for arbitrary concepts; most similar prior work to persona vector pipeline
Neighborhood — ranked by edge-count
Papers (1)
paper
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Benchmark evaluating LLMs as interactive agents in tool-use settings, cited.
- Social deduction game benchmark (Avalon) for LLMs, cited.
- Game-theoretic LLM evaluation benchmark with short-horizon interactions, cited.
- Benchmark for moral understanding in language models; cited as relevant existing evaluation tool
- SkillsBench enforcement mechanism that accepts only single-key JSON actions; composite multi-key actions are rejected, preventing skill loading
- Physical principle of stable states separated by energy barriers, allowing discrete jumps; exemplified in clocks, switches, molecular isomerism, and life.
- The phenomenon where SAEs break a smooth geometric manifold into many small, seemingly unrelated pieces, losing overarching structure.
- A narrower, rarer class of forms that go to the root of human existence, more powerful and awe-filled than mere living structure.