#representation analysis
6 papers match
From Representations to Behaviors: Exploring the Person-Situation-Behavior Triad in LLMs
Ruikang Zhang, Shuo Wang, Qi Su
The paper investigates whether large language models (LLMs) have internal, trait-like personality representations that can be identified, manipulated, and shown to affect behavior…
Prior Directions: Why GUI Grounding Gets Locked in the Past
Weile Gong, Zijian Lu, Mingcai Chen +3
The paper investigates how vision-language models can become locked onto outdated textual priors, causing incorrect visual grounding, and identifies recurring latent directions—cal…
Do Unified Multimodal Models Think in One Space? A Lens Through Cross-Branch Steering
Yu Wang, Sharon Li
The paper investigates whether unified multimodal models share a common semantic space by introducing cross-branch semantic steering, showing that semantic directions from the unde…
Probing the Origins of Reasoning Performance: Representational Quality for Mathematical Problem-Solving in RL vs. SFT Fine-Tuned Models
Antyabha Rahman, Akshaj Gurugubelli, Omar Ankit +2
The paper investigates why reinforcement‑learning‑trained models outperform supervised fine‑tuned models on math reasoning by analyzing their internal representations with linear p…
Do Code Language Models Use Tests? A Behavioral and Representational Study of Test-Driven Code Generation
Yunhao Liang, Chengguang Gan, Ruixuan Ying +3
The paper investigates how code language models respond to test cases in prompts, analyzing whether tests act as executable specifications or merely extra context, and finds that t…
Representation and Reference Selection in Training-Free Synthetic Image Attribution
Meiling Li, Pietro Bongini, Benedetta Tondi +1
The paper investigates how the choice of visual representation and reference selection method affect training‑free, reference‑based attribution of synthetic images, showing that in…
One search, two signals: results blend meaning (embedding similarity, so papers that never use your words still surface) with keyword matches on titles, abstracts and summaries. Free, no sign-in needed.