Latent chain-of-thought replaces written reasoning steps with continuous hidden states, giving models additional steps to compute before answering. It remains unclear what computation these steps actually perform. High task accuracy and decodable intermediate answers do not establish that a model uses its latent chain...
Jing-Jie Qiu, Sheldon Yu, Xun-Yi Jiang et al.· 0 citations
Across verifiable reasoning, open-ended generation, and post-SFT settings, RRPO remains competitive with verifier-based optimization, improves over weakly supervised baselines, and provides additional gains after supervised fine-tuning.
This empirical study investigates whether procedural LLM skills can be represented as directions in activation space and whether vector-space operations over these directions can express skill-level behaviors and finds that procedural skills admit a vector-space representation.
Xun-Yi Jiang, Jun-Da Wu, Yuxin Xiong et al.· 0 citations
Agent2UCB is demonstrated, an agentic GEO system that autonomously improves content visibility through customized, feedback-driven optimization and provides a lightweight, text-only SEO readiness evaluation covering readability, topical coverage, and EEAT-style credibility.
The study turns the filesystem default from an assumption into a design space for agent memory, and turns the model is not the only lever over a store's shape: changing the tool set alone reshapes the store as strongly as swapping the model.
This work treats each reasoning trace as a sequence of latent states rather than an unstructured texts, and investigates whether inference time interventions can provide fine-grained control over the self-looping reasoning process.
