WEZZEL NEWS

A decodability criterion predicts when hidden-state selection beats majority voting in large language models

Researchers have proposed a method called CASE to improve the accuracy of large language models by s...

Toward Personal Intelligence Through Cooperative Observation

Researchers propose a framework called 'cooperative observation' to improve the performance of perso...

KnowSim: Evaluating Information Calibration in LLM Assistants with User Simulators that Learn

Researchers have developed a new framework called KNOWSIM for evaluating the performance of Large La...

Synthesizing Feature Extractors: An Agentic Approach for Algorithm Selection

Researchers have developed an automated approach to designing feature extractors for constraint sati...

Benchmarking the Benchmarks: Evaluating Automated Safety Benchmarks for Small Language Models

Researchers evaluated five widely used benchmark suites on 26 open-source small ...

Fool's Gold: Defensive Deception Against Safety-Removal Attacks on Open-Weight Models

Researchers have developed a new defense mechanism called 'Fool's Gold' to prote...

Explicit State Elicitation Is Not Enough: A Controlled Audit of Memory-Policy Classification

Researchers have developed an audit protocol to test the performance of personal...

Do LLMs Know a Good Hypothesis When They See One? Logit-Based Energy Scoring Outperforms Prompted LLM-as-Judge for Scientific Hypothesis Ranking

Researchers proposed a new method to evaluate scientific hypotheses generated by...