Running on Zero Agents Dialectical Intuitions π€ Judge intuitions; Qwen3-8B writes the view behind them
Sleeping Agents Dialectical Transition Operator v2 π Iterative answer+critique self-improvement (graded RL)
Sleeping Agents Dialectical Transition Operator π Iterative answer + critique self-improvement (Qwen3-8B, RL)
Sleeping Agents Dialectical LLM β Qwen3-32B positions-only RL πΏ Qwen3-32B positions-only RL β distinct answers, no counters
Sleeping Agents Dialectical LLM β Qwen3-32B SFT π³ Qwen3-32B (SFT) β distinct positions + counters as a tree