We have 8 of 52 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
Test-Time Reinforcement Learning (TTRL) enables large language models to adapt directly on unlabeled test problems by constructing pseudo-labels from multiple sampled responses. However, existing TTRL methods typically apply the same expensive sampling and optimization procedure to all test problems, despite substantia...
Reusable Latent Correction (RLC) is proposed, which converts one-off natural-language guidance from a black-box LLM into persistent corrective experiences in the hidden space of an SLM, enabling the SLM to reuse LLM-derived corrections during inference without any online LLM calls.
Learner simulation aims to reproduce how a particular learner behaves on new tasks. Although Large Language Models (LLMs) can generate increasingly fine-grained learning behaviors, existing approaches often need to repeatedly process a growing interaction history to reconstruct the learner. This introduces additional c...
DuoOPD is introduced, in which the student's outcome sets the direction of feedback and the joint teacher-student outcome decides how the teacher supports it: when only the teacher succeeds, its verified answer becomes context for scoring the student's failed response.
This work proposes Step-wise Training for In-context Reasoning (STIR), a model to dynamically decide when to retrieve a single logically consistent next step, just using the current problem and its intermediate state as the query.
A unified reasoning reward model supporting multiple languages and evaluation paradigms, UniRRM uses a staged reasoning chain to dynamically generate task-generic and instruction-specific criteria, enabling fine-grained, input-adaptive judgments while maintaining consistency across languages.
SafeIMG is introduced, a safety-oriented benchmark spanning 12 public- and individual-safety scenarios generated using GPT Image 2.0 that provides human annotations that localise suspicious regions and explain local artefacts and higher-level commonsense or physical inconsistencies.
This work proposes Step-wise Training for In-context Reasoning (STIR), a model to dynamically decide when to retrieve a single logically consistent next step, just using the current problem and its intermediate state as the query.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
