9 papers
Visible but Not Avoided: State-Dependent Re-Proposal of Rejected Candidates in Catalog-Based LLM Optimization Loops
Jianlin Chen, Wenhui Chen, Chi Man Vong
LLM optimization loops often show the proposer the candidates it has already tried and rejected, but leave it to the proposer to avoid them. We ask whether avoidance measured in a…
Self-Cleaning and Captured Anyway: One Measured Primitive for Error in a Store an Agent Writes to Itself, and What a Falling Score Actually Measures
Wenhui Chen, Jianlin Chen, Ziyao Lin +1
"An agent that writes its conclusions into a store it later retrieves from closes a loop usually reported as one-way contamination. Taking the loop to the infinite-tenure limit aga…
The Architect, the Adversary, and the Judge: Closed-Loop Generation of Standards-Aligned Assessment Items at Scale
Wenhui Chen, Ziyao Lin, Jianlin Chen +2
We present CLAIM, a production pipeline for K-12 assessment-item generation coupling a two-stage generate-then-attack protocol (the model drafts as a "curriculum architect", then r…
A Judge Should Know What Changed:Construct Validity for LLM-as-a-Judge Evaluation
Jianlin Chen, Wenhui Chen, Ziyao Lin +1
LLM-as-a-judge evaluation is usually assessed by agreement and robustness to surface perturbations, but reliability does not establish construct validity. We formalize construct va…
Budget-Constrained Embodied Perception: Four Resource Walls and a Pre-Registered Evaluation of Access-Structured Perception on Open Models at less than 31B
Defu Lin, Wenhui Chen, Ziyao Lin +3
Embodied multimodal agents must answer from growing observation streams under a fixed per-decision token budget. We formalize this constraint through four resource walls: a percept…
The Free-Recipe Limit: Every Recipe Effect Measures Which Premise of an Idealised Learner Broke
Wenhui Chen, Jianlin Chen, Ziyao Lin +1
Fix a corpus and send recipe search to infinity: try every order of the skills, every arrangement from blocked to interleaved, every composition, and keep the best. Two quantities…
Judging Is Not Enumerating: Silent Omissions in LLM-Authored Acceptable Sets
Wenhui Chen, Jianlin Chen, Ziyao Lin +2
Language models are increasingly promoted from examinees to examiners: they write the test suites, answer keys, rubrics, and reward functions that define correctness for other syst…
Auditing Discovery Claims: A Two-Sided Criterion for Agentic Science, with the Negative Side Decidable
Wenhui Chen, Jianlin Chen, Ziyao Lin +1
When a self-improving AI-for-science system claims a new capability, the evidence is usually a benchmark delta, a description-length gate, or a p-value. None separates a real gain…
The Capability Convergence Hypothesis: Capability from Access Structure, Not Scale
Wenhui Chen, Jianlin Chen, Ziyao Lin +1
The Platonic Representation Hypothesis (PRH) holds that as models scale, representations of heterogeneous networks converge toward a shared model of reality. We propose its sequel…