64 accepted papers
In this paper, we propose **AlphaGRPO**, a novel framework that applies Group Relative Policy Optimization (GRPO) to AR-Diffusion Unified Multimodal Models (UMMs) to enhance multimodal generation capabilities without relying on external knowledge injection. Our approach unlocks the model's intrinsic…
Recent advances in Vision-Language Models (VLMs) have propelled embodied agents by enabling direct perception, reasoning, and planning task-oriented actions from visual inputs. However, such vision-driven embodied agents open a new attack surface: visual backdoor attacks, where the agent behaves no…
Large Multimodal Models (LMMs) face notable challenges when encountering multimodal knowledge conflicts, particularly under retrieval-augmented generation (RAG) frameworks, where the contextual information from external sources may contradict the model’s internal parametric knowledge, leading to unr
GUI agents are rapidly shifting from multi-module pipelines to end-to-end, native vision-language models (VLMs) that perceive raw screenshots and directly interact with digital devices. Despite rapid progress on general GUI tasks, CAPTCHA solving remains a major challenge. On the other hand, althoug…
Accurate trajectory prediction and motion planning are crucial for autonomous driving systems to navigate safely in complex, interactive environments characterized by multimodal uncertainties. However, current generation-then-evaluation frameworks typically construct multiple plausible trajectory hy…
Although multi-modal large language models (MLLMs) have shown strong capabilities across diverse domains, their application in generating fine-grained 3D perception and prediction outputs in autonomous driving remains underexplored. In this paper, we propose DrivePI, a novel spatial-aware 4D MLLM th
Single-domain offline reinforcement learning (RL) often suffers from limited data coverage, while cross-domain offline RL handles this issue by leveraging additional data from other domains with dynamics shifts. However, existing studies primarily focus on train-time robustness (handling dynamics sh…
In semi-supervised semantic segmentation (SSSS), segmentation performance is heavily constrained by the quality of pseudo labels. However, prevalent pseudo-label optimization approaches rely on the model’s internal self-correction. When the model fails to recognize or adequately represent certain cl
Offline-to-online reinforcement learning (O2O RL) aims to improve the performance of offline pretrained agents through online fine-tuning. Existing O2O RL methods have achieved advances in mitigating the overestimation of Q-value biases (i.e., biases of cumulative rewards), improving the performance…
Sign language is vital for communication within the deaf and hard-of-hearing community. Deep learning techniques, such as Generative Adversarial Networks and diffusion models, have shown promise in generating sign language video content. However, existing methods typically focus on 2D sign language…
We explore the problem of path planning for mobile robots in unknown environments using deep reinforcement learning (DRL). This paper proposes a multi-layer long short-term memory twin delayed deep deterministic policy gradient (MLTD3) algorithm for unknown environments. First, we introduce multi-la
Real-world datasets collected from sensors or human inputs are prone to noise and errors, posing significant challenges for applying offline reinforcement learning (RL). While existing methods have made progress in addressing corrupted actions and rewards, they remain insufficient for handling corru…
The rapid advancements in Vision-Language Models (VLMs) have shown great potential in tackling mathematical reasoning tasks that involve visual context. Unlike humans who can reliably apply solution steps to similar problems with minor modifications, we found that state-of-the-art VLMs like GPT-4o c…
Leveraging Multi-modal Large Language Models (MLLMs) to create embodied agents offers a promising avenue for tackling real-world tasks. While language-centric embodied agents have garnered substantial attention, MLLM-based embodied agents remain underexplored due to the lack of comprehensive evaluat…
With the massive growth of multi-modal information such as text, images, and other data, how should we analyze and align these data becomes very important. In our work, we introduce a new framework based on Reinforcement Learning Guided Graph Diffusion to address the complexity of multi-modal graphs…
One of the principal challenges in building VLM-powered GUI agents is visual grounding—localizing the appropriate screen region for action execution based on both the visual content and the textual plans. Most existing work formulates this as a text-based coordinate generation task. However, these a…
Large language models (LLMs) are widely used, but they often generate subtle factual errors, especially in long-form text. These errors are fatal in some specialized domains such as medicine. Existing fact-checking with grounding documents methods face two main challenges: (1) they struggle to under…
Recent advancements in large language models (LLMs) have significantly propelled the development of large multi-modal models (LMMs), highlighting the potential for general and intelligent assistants. However, most LMMs model visual and textual modalities separately, leading to recent efforts to deve…
Generalization in visual reinforcement learning (VRL) aims to learn agents that can adapt to test environments with unseen visual distractions. Despite advances in robust representations learning, many methods do not take into account the essential downstream task of sequential decision-making. This…
Recent advances have highlighted the benefits of scaling language models to enhance performance across a wide range of NLP tasks. However, these approaches still face limitations in effectiveness and efficiency when applied to domain-specific tasks, particularly for small edge-side models. We propos…
Existing large language model (LLM) evaluation benchmarks primarily focus on English, while current multilingual tasks lack parallel questions that specifically assess cross-lingual reasoning abilities. This dual limitation makes it challenging to assess LLMs’ performance in the multilingual setting
Humans rely on multisensory integration to perceive spatial environments, where auditory cues enable sound source localization in three-dimensional space. Despite the critical role of spatial audio in immersive technologies such as VR/AR, most existing multimodal datasets provide only monaural audi…
Model merging provides a scalable alternative to multi-task training by combining specialized finetuned models through parameter arithmetic, enabling efficient deployment without the need for joint training or access to all task data. While recent methods have shown promise, existing evaluations are…
Reward modeling is a key step in building safe foundation models when applying reinforcement learning from human feedback (RLHF) to align Large Language Models (LLMs). However, reward modeling based on the Bradley-Terry (BT) model assumes a global reward function, failing to capture the inherently d
Generating sketches that accurately reflect the content of reference images presents numerous challenges. Current methods either require paired training data or fail to accommodate a wider range and diversity of sketch styles. While pre-trained diffusion models have shown strong text-based control c…
Deep learning methods have demonstrated remarkable performance across various communication signal processing tasks. However, most signal classification methods require a substantial amount of labeled samples for training, posing significant challenges in the field of communication signals, as label
This paper proposes a novel Q-learning-based dual-loop force tracking control framework for robot grinding tasks in uncertain environments. A complete system state-space model is established, incorporating interaction dynamics and the desired force. By augmenting the system state, a discount cost fu
Quantum computing is an emerging field recognized for the significant speedup it offers over classical computing through quantum algorithms. However, designing and implementing quantum algorithms pose challenges due to the complex nature of quantum mechanics and the necessity for precise control ove…
Recently, many studies focus on utilizing large language models (LLMs) into educational dialogues. Especially, within liberal arts dialogues, educators must balance Humanized communication, Teaching expertise, and Safety-ethics (HTS), besides the subject knowledge itself. However, due to collecting…
Traditional image-to-3D models often struggle with scenes containing multipleobjects due to biases and occlusion complexities. To address this challenge, wepresent REPARO, a novel approach for compositional 3D asset generation fromsingle images. REPARO employs a two-step process: first, it extracts…
Multi-hop question answering (QA) remains challenging, as solutions must reliably integrate and reconcile evidence from multiple sources without succumbing to error propagation. While large language models (LLMs) have achieved substantial improvements via chain-of-thought (CoT) prompting and retriev
Understanding human preferences is crucial for improving foundation models and building personalized AI systems. However, preferences are inherently diverse and complex, making it difficult for traditional reward models to capture their full range. While fine-grained preference data can help, collec…
Generating sketches guided by reference styles requires precise transfer of stroke attributes, such as line thickness, deformation, and texture sparsity, while preserving semantic structure and content fidelity. To this end, we propose Stroke2Sketch, a novel training-free framework that introduces c…
Learning policy from offline datasets through offline reinforcement learning (RL) holds promise for scaling data-driven decision-making while avoiding unsafe and costly online interactions. However, real-world data collected from sensors or humans often contains noise and errors, posing a significan…
Audio temporal forgery localization (ATFL) aims to find the precise forgery regions of the partial spoof audio that is purposefully modified. Existing ATFL methods rely on training efficient networks using fine-grained annotations, which are obtained costly and challenging in real-world scenarios. T
Fine-grained control over large language models (LLMs) remains a significant challenge, hindering their adaptability to diverse user needs. While Reinforcement Learning from Human Feedback (RLHF) shows promise in aligning LLMs, its reliance on scalar rewards often limits its ability to capture diver…
With the development of AI-Generated Content (AIGC), text-to-audio models are gaining widespread attention. However, it is challenging for these models to generate audio aligned with human preference due to the inherent information density of natural language and limited model understanding ability.…
Diffusion models have made significant strides in visual content generation but also raised increasing demands on generated image detection. Existing detection methods have achieved considerable progress, but they usually suffer a significant decline in accuracy when detecting images generated by an…
Educational materials such as survey articles in specialized fields like computer science traditionally require tremendous expert inputs and are therefore expensive to create and update. Recently, Large Language Models (LLMs) have achieved significant success across various general tasks. However, t…
In recent years, multi-agent path finding (MAPF) has attracted widespread attention in the fields of artificial intelligence and robotics. Its main goal is to find paths for multiple agents, each having specified start and end locations on a grid, without collisions, while minimizing the total trave…
Autonomous vehicles require online learning capabilities to enable long-term, unattended operation. However, long-term online learning is accompanied by the problem of forgetting previously learned knowledge. This paper introduces an online learning framework that includes a catastrophic forgetting…
We present a Multi-Stereoscopic Visual-Inertial Odometry (VIO) system capable of integrating an arbitrary number of stereo cameras, exhibiting excellent robustness in the face of visually challenging scenarios. During system initialization, we introduce multi-view keyframes for simultaneous processi
Reward models trained on human preference data have been proven to effectively align Large Language Models (LLMs) with human intent within the framework of reinforcement learning from human feedback (RLHF). However, current reward models have limited generalization capabilities to unseen prompts and…
We consider the problem of multi-objective alignment of foundation models with human preferences, which is a critical step towards helpful and harmless AI systems. However, it is generally costly and unstable to fine-tune large foundation models using reinforcement learning (RL), and the multi-dimen…
Offline reinforcement learning (RL) presents a promising approach for learning reinforced policies from offline datasets without the need for costly or unsafe interactions with the environment. However, datasets collected by humans in real-world environments are often noisy and may even be malicious…
Real-world offline datasets are often subject to data corruptions (such as noise or adversarial attacks) due to sensor failures or malicious attacks. Despite advances in robust offline reinforcement learning (RL), existing methods struggle to learn robust agents under high uncertainty caused by the…
Box-supervised instance segmentation has gained much attention as it requires only simple box annotations instead of costly mask or polygon annotations. However, existing box-supervised instance segmentation models mainly focus on mask-based frameworks. We propose a new end-to-end training technique…
We investigate the problem of corruption robustness in offline reinforcement learning (RL) with general function approximation, where an adversary can corrupt each sample in the offline dataset, and the corruption level $\zeta\geq0$ quantifies the cumulative corruption amount over $n$ episodes and $…
This paper aims to efficiently enable Large Language Models (LLMs) to use multi-modal tools. The advanced proprietary LLMs, such as ChatGPT and GPT-4, have shown great potential for tool usage through sophisticated prompt engineering. Nevertheless, these models typically rely on prohibitive computat…
Domain adaptive semantic segmentation aims to adapt a model trained on labeled source domain to the unlabeled target domain. Self-training shows competitive potential in this field. Existing methods along this stream mainly focus on selecting reliable predictions on target data as pseudo-labels for…
Recent work has shown that representation learning plays a critical role in sample-efficient reinforcement learning (RL) from pixels. Unfortunately, in real-world scenarios, representation learning is usually fragile to task-irrelevant distractions such as variations in background or viewpoint. To t…
Interactive devices with touch screen have become commonly used in various aspects of daily life, which raises the demand for high production quality of touch screen glass. While it is desirable to develop effective defect detection technologies to optimize the automatic touch screen production line…
While deep reinforcement learning (RL) has been demonstrated effective in solving complex control tasks, sample efficiency remains a key challenge due to the large amounts of data required for remarkable performance. Existing research explores the application of representation learning for data-effi…
Offline goal-conditioned RL (GCRL) offers a way to train general-purpose agents from fully offline datasets. In addition to being conservative within the dataset, the generalization ability to achieve unseen goals is another fundamental challenge for offline GCRL. However, to the best of our knowled…
In this work, we study the simple yet universally applicable case of reward shaping in value-based Deep Reinforcement Learning (DRL). We show that reward shifting in the form of a linear transformation is equivalent to changing the initialization of the $Q$-function in function approximation. Based…
Offline reinforcement learning (RL) provides a promising direction to exploit massive amount of offline data for complex decision-making tasks. Due to the distribution shift issue, current offline RL algorithms are generally designed to be conservative in value estimation and action selection. Howev…
Solving goal-conditioned tasks with sparse rewards using self-supervised learning is promising because of its simplicity and stability over current reinforcement learning (RL) algorithms. A recent work, called Goal-Conditioned Supervised Learning (GCSL), provides a new learning framework by iterativ…
Learning-based methods are powerful in handling complex scenarios. However, it is still challenging to use learning-based methods under uncertain environments while stability, safety, and real-time performance of the system are desired to guarantee. In this letter, we propose a learning-based tracki
"The vanilla self-attention mechanism inherently relies on pre-defined and steadfast computational dimensions. Such inflexibility restricts it from possessing context-oriented generalization that can bring more contextual cues and graphic representations. To mitigate this issue, we propose a Scalabl…
We study the offline meta-reinforcement learning (OMRL) problem, a paradigm which enables reinforcement learning (RL) algorithms to quickly adapt to unseen tasks without any interactions with the environments, making RL truly practical in many real-world applications. This problem is still not fully…
Fact verification is a challenging task which requires to retrieve relevant sentences from plain texts and then take these sentences as evidences to verify given claims. Conventional methods treat sentence selection and claim verification as separate subtasks in a pipeline. Claim verification models…
Accurate path following is challenging for autonomous robots operating in uncertain environments. Adaptive and predictive control strategies are crucial for a nonlinear robotic system to achieve high-performance path following control. In this letter, we propose a novel learning-based predictive con
We introduce a Multiple Granularity Analysis framework for video segmentation in a coarse-to-fine manner. We cast video segmentation as a spatio-temporal superpixel labeling problem. Benefited from the bounding volume provided by off-the-shelf object trackers, we estimate the foreground/ background…
The widespread availability of audio editing software has made it very easy to create forgeries without perceptual trace. Copy-move is one of popular audio forgeries. It is very important to identify audio recording with duplicated segments. However, copy-move detection in digital audio with sample…