LPSR raises 8B-model accuracy on MATH-500 from 28.8% to 44.0% by detecting error-indicating phase shifts in the residual stream and correcting via KV-cache rollback plus steering vectors, outperforming prompted self-correction and even a 70B model.
Tree of thoughts: Deliberate problem solving with large language models
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Hubble is an LLM-driven framework that safely discovers diverse alpha factors via operator trees, RAG feedback, and out-of-sample validation on US equity data, with range and volatility factors showing persistence.
citing papers explorer
-
Latent Phase-Shift Rollback: Inference-Time Error Correction via Residual Stream Monitoring and KV-Cache Steering
LPSR raises 8B-model accuracy on MATH-500 from 28.8% to 44.0% by detecting error-indicating phase shifts in the residual stream and correcting via KV-cache rollback plus steering vectors, outperforming prompted self-correction and even a 70B model.
-
Hubble: An LLM-Driven Agentic Framework for Safe, Diverse, and Reproducible Alpha Factor Discovery
Hubble is an LLM-driven framework that safely discovers diverse alpha factors via operator trees, RAG feedback, and out-of-sample validation on US equity data, with range and volatility factors showing persistence.