Audio-visual speech enhancement (AVSE) leverages visual cues to recover target speech from mixtures corrupted by reverberation, noise, and overlapping speakers. We propose a coarse-to-fine two-stage AVSE framework. In the first stage, a progressive SPAV-TFGridNet is extended to two synchronized visual streams to produc...
Long-Jie Luo, Zhong-Shu Hou, Yue-Peng Li et al.· 0 citations
Artificial intelligence systems deployed in high-risk settings increasingly require models that support meaningful human oversight. Concept-based models (CMs) offer a promising framework by enabling users to inspect and intervene on intermediate, human-understandable concepts. However, concept interventions can counter...
Gregory Verghese, Christopher J. Soelistyo, Mateo Espinosa Zarlenga et al.· 0 citations
Many collective-action systems are neither fully decentralized nor fully centralized: agents voluntarily organize into institutions that share organizational costs and delegate decisions to representatives. Delegation can reduce individual costs, but it also transfers control to representatives whose private objectives...
A. Mukhopadhyay, Aris Filos-Ratsikas, Panayiotis Danassis· 0 citations
As agents powered by foundation models move into open-ended environments, continual learning from experience becomes essential. For agents built on models with fixed parameters, this learning requires reusing prior task experience in context to improve performance and efficiency on future tasks. Existing memory and exp...
Yi-Xu Huang, Jia-Qi W. Ma· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Web search allows LLMs to access up-to-date information, but it remains unclear why they sometimes produce outdated answers even when relevant evidence is available. To study this problem, we construct temporal facts whose values changed after the models' training cutoffs. We evaluate one proprietary model under deploy...
Federated LLM unlearning addresses targeted knowledge removal when forget and retain data remain distributed across private clients and are inaccessible to the server. The resulting locally trained adapters encode conflicting objectives, making their aggregation fundamentally different from conventional federated optim...
Reeyansh Shah, Maram Hasan, Subhasis Chaudhuri et al.· 0 citations
Language embeddings are increasingly used in recommenders to represent item content, but their privacy risks remain poorly understood. We study whether item embeddings can reveal item identities when an adversary has access to auxiliary item text from external sources. We show that a simple cosine-similarity attack can...
When an LLM agent misbehaves, a human steps in, and that intervention is usually discarded or replayed as training data. We treat it instead as a measurement: a label pointing at the internal computation that failed. We study it in a controlled agent environment, on one failure (asking a clarifying question when the re...
Cesare Zavattari, Alessandro Tommasi, Giuseppe Prencipe· 0 citations
Reinforcement learning (RL) can become off-policy due to training-inference mismatch, mini-batch updates, and asynchronous training.The resulting mismatch is especially harmful in Mixture-of-Experts (MoE) models, where rollout--training discrepancies can change routing decisions and lead to router drift. Rollout Routin...
Liangyao Jin, Zhixin Wang, Liming Liu et al.· 0 citations
Experiments across nearly 200 benchmark classification datasets using Decision Tree, Multilayer Perceptron, Gaussian Naive Bayes, and Support Vector Classifier as base models reveal that the proposed triplet with absolute difference encoder achieves the highest average F1 scores.
Bernard Georges, Mohamed Karim Belaid, Manuel Rodrigues· WorldCIST· 0 citations
Large audio language models (LALMs) have made substantial progress in audio understanding. However, autoregressive text generation introduces computational overhead and latency in high-throughput applications that require frequent selection, judgment, and scoring. To address this limitation, we introduce Qwen-Audio-Jev...
Pei-Xin Chen, Zhi-Gao Chen, Zhi-Yun Fan et al.· 0 citations
Reverse engineering the data formats used by mobile software is skilled and slow, and using AI to assist with it is attractive. We evaluate this for the protobuf messages that an iPhone sends to Apple's location services. Working from Apple's binaries, we use two frontier models (Claude Opus 4.8, Codex GPT Sol 5.6) to...
