Deploying generalist robotic agents in the real world requires transferable skills. Specifically, a policy trained to clone a behavior from object-specific demonstrations must generalize beyond that object, otherwise data collection requirements become intractable. Recently, fine-tuning of pre-trained billion-parameter...
Nikolaos Tsagkas, Andreas Sochopoulos, Chris Xiaoxuan Lu et al.· 0 citations
Runtime monitors for autonomous agents commonly threshold an accumulated internal state - a behavioural baseline, a drift statistic, or, in our prior work, a modelled affective state. We previously reported a State Saturation Trap: threshold-on-state triggers over a continuous affect engine become near-constant alarms...
A prominent research direction in mechanistic interpretability is learning sparse circuits over LLM components to reveal how they jointly produce model behavior. However, raw neurons are polysemantic, making learned circuits hard to interpret. Sparse autoencoder (SAE) features alleviate this, but their high dimensional...
Naiyu Yin, Dennis Wei, Tian Gao et al.· 0 citations
Multi-turn tool-use agents must reason, call tools, and adapt to observations across several interaction turns. Post-training such agents is challenging, as reinforcement learning often suffers from sparse rewards and weak credit assignment despite matching the prompt-only inference setting, while supervised fine-tunin...
Zhenbang Du, Jun Luo, Zhiwei Zheng et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Objective: To enhance the accuracy, interpretability, and robustness of large language models (LLMs) in medical question answering (MedQA).
Method: We designed a multi-agent peer-reviewed reasoning method in which multiple LLM agents independently generate chain-of-thought reasoning with candidate answers, then act a...
Modern video diffusion models generate increasingly realistic and temporally coherent videos, motivating their use as candidate world simulators. Yet it remains unclear whether these models internally encode physical structure, or merely reproduce motion patterns seen during training. We study this question by probing...
Parsa Esmati, Somjit Nath, Katja Hofmann et al.· 0 citations
The identification of constitutive neural network models from heterogeneous full-field deformation data provides a robust alternative to traditional calibration methods based on homogeneous stress-strain experiments, particularly given the high dimensionality of trainable parameters. This work presents ADE-FEMU (Automa...
Matthias Knipper, Chenyi Ji, Malte Brand et al.· 0 citations
Self on-policy distillation trains a student policy against a teacher derived from its own parameter history, yet the teacher's update schedule -- which governs the \emph{temporal coupling} between teacher and student -- has not been systematically studied as a stability variable. Through a controlled schedule sweep on...
Haowei Guo, Baolong Bi, Ruicheng Zhang et al.· 0 citations
Electromyography (EMG) records muscle activation for hand gesture recognition, prosthetic control, and wearable interaction. Existing EMG methods usually formulate hand action understanding as prediction over fixed gesture labels, leaving EMG signals and action-semantic text unaligned in a common feature space and limi...
Latent reasoning enables reasoning over continuous hidden states rather than explicit tokens, avoiding the language bottleneck and inference overhead of chain-of-thought for medical VQA. However, existing methods suffer from modality collapse, insufficient visual supervision, and train-inference mismatch. Moreover, the...
Qiaoru Li, Shaotian Liang, Jintao Chen et al.· 0 citations
Safe exploration remains a fundamental challenge in reinforcement learning (RL), limiting the deployment of RL agents in the real world. We propose Sampling-Based Safe Reinforcement Learning (SBSRL), a model-based RL algorithm that maintains safety throughout the learning process by enforcing constraints jointly across...
Luca Vignola, Bruce D. Lee, Manish Prajapat et al.· 0 citations
