Related papers: TorR: Towards Brain-Inspired Task-Oriented Reasoni…

Tool-Integrated Reasoning (TIR) has significantly enhanced the capabilities of Large Language Models (LLMs), yet current agents tend to exhibit cognitive offloading, redundantly invoking external tools even for simple tasks. In this paper,…

Contrastive image-text models such as CLIP form the building blocks of many state-of-the-art systems. While they excel at recognizing common generic concepts, they still struggle on fine-grained entities which are rare, or even absent from…

While attention has been an increasingly popular component in deep neural networks to both interpret and boost performance of models, little work has examined how attention progresses to accomplish a task and whether it is reasonable. In…

Traditional machine learning depends on high-precision arithmetic and near-ideal hardware assumptions, which is increasingly challenged by variability in aggressively scaled semiconductor devices. Compute-in-memory (CIM) architectures…

Reinforcement learning (RL) with verifiable rewards has proven effective at post-training LLMs for coding, yet deploying separate task-specific specialists incurs costs that scale with the number of tasks, motivating a unified multi-task RL…

Diffusion language models hold the promise of fast parallel generation, while autoregressive (AR) models typically excel in quality due to their causal structure aligning naturally with language modeling. This raises a fundamental question:…

Composed Image Retrieval (CIR) aims to retrieve a target image based on a reference image and conditioning text, enabling controllable image searches. The mainstream Zero-Shot (ZS) CIR methods bypass the need for expensive training CIR…

Future networks (including 6G) are poised to accelerate the realisation of Internet of Everything. However, it will result in a high demand for computing resources to support new services. Mobile Edge Computing (MEC) is a promising…

Cooperative perception via communication among intelligent traffic agents has great potential to improve the safety of autonomous driving. However, limited communication bandwidth, localization errors and asynchronized capturing time of…

This thesis provides an in-depth structural analysis and efficient algorithmic solutions for tabletop object rearrangement with overhand grasps (TORO), a foundational task in advancing intelligent robotic manipulation. Rearranging multiple…

Autoregressive image generation has seen recent improvements with the introduction of chain-of-thought and reinforcement learning. However, current methods merely specify "What" details to depict by rewriting the input prompt, yet…

Multimodal large language models are increasingly expected to perform thinking with images, yet existing visual latent reasoning methods still rely on explicit textual chain-of-thought interleaved with visual latent tokens. This interleaved…

Large language models can exhibit emergent reasoning behaviors, often manifested as recurring lexical patterns (e.g., "wait," indicating verification). However, complex reasoning trajectories remain sparse in unconstrained sampling, and…

Text embedding and generative tasks are usually trained separately based on large language models (LLMs) nowadays. This causes a large amount of training cost and deployment effort. Context compression is also a challenging and pressing…

The Transformer architecture excels in a variety of language modeling tasks, outperforming traditional neural architectures such as RNN and LSTM. This is partially due to its elimination of recurrent connections, which allows for parallel…

Video Reasoning Segmentation (VRS) aims to segment target objects in videos based on implicit instructions that convey human intent and temporal logic. Existing MLLM-based methods predict masks with a [SEG] token after selecting frames via…

Machine learning has enabled the use of implicit neural representations (INRs) to efficiently compress and reconstruct massive scientific datasets. However, despite advances in fast INR rendering algorithms, INR-based rendering remains…

Graph-based Retrieval-Augmented Generation (GraphRAG) has become the important paradigm for enhancing Large Language Models (LLMs) with external knowledge. However, existing approaches are constrained by their reliance on high-quality…

Assessing the quality of outputs generated by generative models, such as large language models and vision language models, presents notable challenges. Traditional methods for evaluation typically rely on either human assessments, which are…

Graph Retrieval-Augmented Generation (GraphRAG) is dominated by a retrieve-then-reason paradigm, where context is retrieved using heuristics and then reasoned over. Such methods struggle to adapt to the query-specific logic required for…