In open-vocabulary mobile manipulation (OVMM), task success often hinges on the selection of an appropriate base placement for the robot. Existing approaches typically navigate to proximity-based regions without considering affordances, resulting in frequent manipulation failures. We propose Affordance-Guided Coarse-to-Fine Exploration, a zero-shot framework for base placement that integrates semantic understanding from vision-language models (VLMs) with geometric feasibility through an iterative optimization process. Our method constructs cross-modal representations, namely Affordance RGB and Obstacle Map+, to align semantics with spatial context. This enables reasoning that extends beyond the egocentric limitations of RGB perception. To ensure interaction is guided by task-relevant affordances, we leverage coarse semantic priors from VLMs to guide the search toward task-relevant regions and refine placements with geometric constraints, thereby reducing the risk of convergence to local optima. Evaluated on five diverse open-vocabulary mobile manipulation tasks, our system achieves an 85% success rate, significantly outperforming classical geometric planners and VLM-based methods. This demonstrates the promise of affordance-aware and multimodal reasoning for generalizable, instruction-conditioned planning in OVMM.

Pei-An Chen, Yongching Liang, Jia-Fong Yeh, Hung-Ting Su, Yi-Ting Chen, Min Sun, Winston H. Hsu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Conventional Vision-and-Language Navigation (VLN) benchmarks assume instructions are feasible and the referenced target exists, leaving agents ill-equipped to handle false-premise goals.We introduce VLN-NF, a benchmark with false-premise instructions where the target is absent from the specified room and agents must navigate, gather evidence through in-room exploration, and explicitly output NOT-FOUND.VLN-NF is constructed via a scalable pipeline that rewrites VLN instructions using an LLM and verifies target absence with a VLM, producing plausible yet factually incorrect goals.We further propose REV-SPL to jointly evaluate room reaching, exploration coverage, and decision correctness.To address this challenge, we present ROAM, a two-stage hybrid that combines supervised room-level navigation with LLM/VLM-driven in-room exploration guided by a free-space clearance prior.ROAM achieves the best REV-SPL among compared methods, while baselines often under-explore and terminate prematurely under unreliable instructions.

Gueter Josmy Faure, Min-Hung Chen, Jia-Fong Yeh, Ying Cheng, Hung-Ting Su, Yung-Hao Tang, Shang-Hong Lai, Winston H. Hsu. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Long-form video understanding presents unique challenges that extend beyond traditional short-video analysis approaches, particularly in capturing long-range dependencies, processing redundant information efficiently, and extracting high-level semantic concepts. To address these challenges, we propose a novel approach that more accurately reflects human cognition. This paper introduces HERMES: temporal-coHERent long-forM understanding with Episodes and Semantics, featuring two versatile modules that can enhance existing video-language models or operate as a standalone system. Our Episodic COmpressor (ECO) efficiently aggregates representations from micro to semi-macro levels, reducing computational overhead while preserving temporal dependencies. Our Semantics ReTRiever (SeTR) enriches these representations with semantic information by focusing on broader context, dramatically reducing feature dimensionality while preserving relevant macro-level information. We demonstrate that these modules can be seamlessly integrated into existing SOTA models, consistently improving their performance while reducing inference latency by up to 43% and memory usage by 46%. As a standalone system, HERMES achieves state-of-the-art performance across multiple long-video understanding benchmarks in both zero-shot and fully-supervised settings.

To address the limitations of traffic prediction from location-bound detectors, we present Geographical Cellular Traffic (GCT) flow, a novel data source that leverages the extensive coverage of cellular traffic to capture mobility patterns. Our extensive analysis validates its potential for transportation. Focusing on vehicle-related GCT flow prediction, we propose a graph neural network that integrates multivariate, temporal, and spatial facets for improved accuracy. Experiments reveal our model's superiority over baselines, especially in long-term predictions. We also highlight the potential for GCT flow integration into transportation systems.

We introduce a new task called Adaptable Error Detection (AED), which aims to identify behavior errors in few-shot imitation (FSI) policies based on visual observations in novel environments. The potential to cause serious damage to surrounding areas limits the application of FSI policies in real-world scenarios. Thus, a robust system is necessary to notify operators when FSI policies are inconsistent with the intent of demonstrations. This task introduces three challenges: (1) detecting behavior errors in novel environments, (2) identifying behavior errors that occur without revealing notable changes, and (3) lacking complete temporal information of the rollout due to the necessity of online detection. However, the existing benchmarks cannot support the development of AED because their tasks do not present all these challenges. To this end, we develop a cross-domain AED benchmark, consisting of 322 base and 153 novel environments. Additionally, we propose Pattern Observer (PrObe) to address these challenges. PrObe is equipped with a powerful pattern extractor and guided by novel learning objectives to parse discernible patterns in the policy feature representations of normal or error states. Through our comprehensive evaluation, PrObe demonstrates superior capability to detect errors arising from a wide range of FSI policies, consistently surpassing strong baselines. Moreover, we conduct detailed ablations and a pilot study on error correction to validate the effectiveness of the proposed architecture design and the practicality of the AED task, respectively. The AED project page can be found at https://aed-neurips.github.io/.

Traffic prediction utility often faces infrastructural limitations, which restrict its coverage. To overcome this challenge, we present Geographical Cellular Traffic (GCT) flow that leverages cellular network data as a new source for transportation evaluation. The broad coverage of cellular networks allows GCT flow to capture various mobile user activities across regions, aiding city authorities in resource management through precise predictions. Acknowledging the complexity arising from the diversity of mobile users in GCT flow, we supplement it with camera-based vehicle flow data from limited deployments and verify their spatio-temporal attributes and correlations through extensive data analysis. Our two-stage fusion approach integrates these multi-source data, addressing their coverage and magnitude discrepancies, thereby enhancing the prediction of GCT flow for accurate transportation evaluation. Overall, we propose novel uses of telecom data in transportation and verify its effectiveness in multi-source fusion with vision-based data.

The large amount of data collected by LiDAR sensors brings the issue of LiDAR point cloud compression (PCC). Previous works on LiDAR PCC have used range image representations and followed the predictive coding paradigm to create a basic prototype of a coding framework. However, their prediction methods give an inaccurate result due to the negligence of invalid pixels in range images and the omission of future frames in the time step. Moreover, their handcrafted design of residual coding methods could not fully exploit spatial redundancy. To remedy this, we propose a coding framework BIRD-PCC. Our prediction module is aware of the coordinates of invalid pixels in range images and takes a bidirectional scheme. Also, we introduce a deep-learned residual coding module that can further exploit spatial redundancy within a residual frame. Experiments conducted on SemanticKITTI and KITTI-360 datasets show that BIRD-PCC outperforms other methods in most bitrate conditions and generalizes well to unseen environments.

While recent progress has significantly boosted few-shot classification (FSC) performance, few-shot object detection (FSOD) remains challenging for modern learning systems. Existing FSOD systems follow FSC approaches, ignoring critical issues such as spatial variability and uncertain representations, and consequently result in low performance. Observing this, we propose a novelDual-Awareness Attention (DAnA)mechanism that enables networks to adaptively interpret the given support images. DAnA transforms support images intoquery-position-aware(QPA) features, guiding detection networks precisely by assigning customized support information to each local region of the query. In addition, the proposed DAnA component is flexible and adaptable to multiple existing object detection frameworks. By adopting DAnA, conventional object detection networks, Faster R-CNN and RetinaNet, which are not designed explicitly for few-shot learning, reach state-of-the-art performance in FSOD tasks. In comparison with previous methods, our model significantly increases the performance by 47% (+6.9 AP), showing remarkable ability under various evaluation settings.

In few-shot imitation learning (FSIL), using behavioral cloning (BC) to solve unseen tasks with few expert demonstrations becomes a popular research direction. The following capabilities are essential in robotics applications: (1) Behaving in compound tasks that contain multiple stages. (2) Retrieving knowledge from few length-variant and misalignment demonstrations. (3) Learning from an expert different from the agent. No previous work can achieve these abilities at the same time. In this work, we conduct FSIL problem under the union of above settings and introduce a novel stage conscious attention network (SCAN) to retrieve knowledge from few demonstrations simultaneously. SCAN uses an attention module to identify each stage in length-variant demonstrations. Moreover, it is designed under demonstration-conditioned policy that learns the relationship between experts and agents. Experiment results show that SCAN can perform in complicated compound tasks without fine-tuning and provide the explainable visualization. Project page is at https://sites.google.com/view/scan-aaai2022.

Monocular 3D object detection is an important yet challenging task in autonomous driving. Some existing methods leverage depth information from an off-the-shelf depth estimator to assist 3D detection, but suffer from the additional computational burden and achieve limited performance caused by inaccurate depth priors. To alleviate this, we propose MonoDTR, a novel end-to-end depth-aware transformer network for monocular 3D object detection. It mainly consists of two components: (1) the Depth-Aware Feature Enhancement (DFE) module that implicitly learns depth-aware features with auxiliary supervision without requiring extra computation, and (2) the Depth-Aware Transformer (DTR) module that globally integrates context- and depth-aware features. Moreover, different from conventional pixel-wise positional encodings, we introduce a novel depth positional encoding (DPE) to inject depth positional hints into transformers. Our proposed depth-aware modules can be easily plugged into existing image-only monocular 3D object detectors to improve the performance. Extensive experiments on the KITTI dataset demonstrate that our approach outperforms previous state-of-the-art monocular-based methods and achieves real-time detection. Code is available at https://github.com/kuanchihhuang/.MonoDTR.

Depth estimation is essential for many applications, such as 3D reconstruction and object recognition. In practice, the captured depth images naturally suffer from the noise and invalid depth value, i.e, depth noise and depth hole. Despite making a process on filling a small number (less than 10%) of depth holes, the performance of the prior depth reconstruction systems drops significantly when coping with the larger hole region (up to 25%). To tackle this, we propose a novel Sequential Depth Networks (SeqDNet). Instead of filling holes in coarse scale, our model tends to leverage the information in the spatial structures constructed from depth images. There are two novel components in the proposed SeqDNet: (1) We slice the depth map from near to far, generating the spatial structure, named sequential depth. It provides residual depth information, which was neglected in previous works. (2) The edge awareness loss (EA Loss) matches the gradient of the depth map by controlling the weight with the normal map so as to provide additional instruction (whether the position is the edge or not) when filling depth holes. Experimental results show a significant performance gain on filling images with 25% depth holes using SeqDNet. Moreover, a comprehensive analysis is given to pave a path for future researches addressing the large hole size on the depth maps.

Spatial-temporal prediction is a critical problem for intelligent transportation, which is helpful for tasks such as traffic control and accident prevention. Previous studies rely on large-scale traffic data collected from sensors. However, it is unlikely to deploy sensors in all regions due to the device and maintenance costs. This paper addresses the problem via outdoor cellular traffic distilled from over two billion records per day in a telecom company, because outdoor cellular traffic induced by user mobility is highly related to transportation traffic. We study road intersections in urban and aim to predict future outdoor cellular traffic of all intersections given historic outdoor cellular traffic. Furthermore, we propose a new model for multivariate spatial-temporal prediction, mainly consisting of two extending graph attention networks (GAT). First GAT is used to explore correlations among multivariate cellular traffic. Another GAT leverages the attention mechanism into graph propagation to increase the efficiency of capturing spatial dependency. Experiments show that the proposed model significantly outperforms the state-of-the-art methods on our dataset.

Understanding and comprehending video content is crucial for many real-world applications such as search and recommendation systems. While recent progress of deep learning has boosted performance on various tasks using visual cues, deep cognition to reason intentions, motivation, or causality remains challenging. Existing datasets that aim to examine video reasoning capability focus on visual signals such as actions, objects, relations, or could be answered utilizing text bias. Observing this, we propose a novel task, along with a new dataset: Trope Understanding in Movies and Animations (TrUMAn), with 2423 videos associated with 132 tropes, intending to evaluate and develop learning systems beyond visual signals. Tropes are frequently used storytelling devices for creative works. By coping with the trope understanding task and enabling the deep cognition skills of machines, data mining applications and algorithms could be taken to the next level. To tackle the challenging TrUMAn dataset, we present a Trope Understanding and Storytelling (TrUSt) with a new Conceptual Storyteller module, which guides the video encoder by performing video storytelling on a latent space. Experimental results demonstrate that state-of-the-art learning systems on existing tasks reach only 12.01% of accuracy with raw input signals. Also, even in the oracle case with human-annotated descriptions, BERT contextual embedding achieves at most 28% of accuracy. Our proposed TrUSt boosts the model performance and reaches 13.94% performance. We also provide detailed analysis to pave the way for future research. TrUMAn is publicly available at:https://www.cmlab.csie.ntu.edu.tw/project/trope

Dense depth estimation plays a key role in multiple applications such as robotics, 3D reconstruction, and augmented reality. While sparse signal, e.g., LiDAR and Radar, has been leveraged as guidance for enhancing dense depth estimation, the improvement is limited due to its low density and imbalanced distribution. To maximize the utility from the sparse source, we propose Sparse Signal Superdensity (S3) technique, which expands the depth value from sparse cues while estimating the confidence of expanded region. The proposed S3can be applied to various guided depth estimation approaches and trained end-to-end at different stages, including input, cost volume and output. Extensive experiments demonstrate the effectiveness, robustness, and flexibility of the S3technique on LiDAR and Radar signal.

Multi-party dialogue question answering (MPDQA) is an emerging topic in speech and language processing where the goal is to answer the questions according to the multiparty conversations. Different from conventional QA, which assumes a single speaker (writer) and general listeners (readers), MPDQA involves multiple speakers and specific listeners. Prior works simply treat the dialogues as plain passages, which neglect the importance of role awareness, such as speakers’ perspectives and co-references. In a novel aspect, this paper proposes the Role Aware Multi-Party Network (RAMPNet), a model utilizing the information of speaker and role to present "who is speaking" and "who is mentioned", making role awareness an available message for our model. Experiments show that our RAMPNet outperforms the BERT baseline model on a large-scale MPDQA dataset, FriendsQA, especially in the "Who" and "How" questions, which strongly need the ability of conversation relations understanding to answer these questions. In addition, our further analysis demonstrates RAMPNet’s effectiveness in those questions contain verbs such as "reply" or "talk", related to interactions between speakers and roles. All of the results show the capability and utility of RAMPNet.

Despite the success of deep learning on supervised point cloud semantic segmentation, obtaining large-scale point-by-point manual annotations is still a significant challenge. To reduce the huge annotation burden, we propose a Region-based and Diversity-aware Active Learning (ReDAL), a general framework for many deep learning approaches, aiming to automatically select only informative and diverse sub-scene regions for label acquisition. Observing that only a small portion of annotated regions are sufficient for 3D scene understanding with deep learning, we use softmax entropy, color discontinuity, and structural complexity to measure the information of sub-scene regions. A diversity-aware selection algorithm is also developed to avoid redundant annotations resulting from selecting informative but similar regions in a querying batch. Extensive experiments show that our method highly outperforms previous active learning strategies, and we achieve the performance of 90% fully supervised learning, while less than 15% and 5% annotations are required on S3DIS and SemanticKITTI datasets, respectively.

Ke-Jyun Wang, Yun-Hsuan Liu, Hung-Ting Su, Jen-Wei Wang, Yu-Siang Wang, Winston Hsu, Wen-Chin Chen. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Object counting which aims to calculate the number of total instances of the given class is a classic but crucial task that can be applied to many applications. Most of the prior works only focus on counting certain classes of objects such as people, cars, animals, etc. However, in recent years, there are lots of applications that need to get the count of the unseen class of objects such as a mechanical arm commanded to grab the novel object. In this paper, we present an effective object counting network, Class-agnostic Few-shot Object Counting Network (CFOCNet), that supports counting arbitrary classes of object unseen during training stage. Instead of counting a pre-defined class, our model is able to count instances based on input reference images and reduces the huge cost of data collection, training and parameter tuning for each new object class. Our model utilizes not only the similarity between query image and reference images but self attending the query image to learn the self-repeatedness. Using a two-stream Resnet that matches features in different scales, our network can automatically learn to aggregate different scales of the matching scores. We evaluate our method on the subset of the COCO dataset that contains 80 classes of objects and many diverse scenes. In the experiments, our network outperforms other methods including detection and some previous works by a large margin. To the best of our knowledge, we are the first that mainly focuses on few-shot object counting in the class-agnostic manner.

The human ability of deep cognitive skills is crucial for the development of various real-world applications that process diverse and abundant user generated input. While recent progress of deep learning and natural language processing have enabled learning system to reach human performance on some benchmarks requiring shallow semantics, such human ability still remains challenging for even modern contextual embedding models, as pointed out by many recent studies [9, 10, 22, 24, 32]. Existing machine comprehension datasets assume sentence-level input, lack of casual or motivational inferences, or can be answered with question-answer bias. Here, we present a challenging novel task, trope detection on films, in an effort to create a situation and behavior understanding for machines. Tropes are frequently used storytelling devices for creative works. Comparing to existing movie tag prediction tasks, tropes are more sophisticated as they can vary widely, from a moral concept to a series of circumstances, and embedded with motivations and cause-and-effects. We introduce a new dataset, Tropes in Movie Synopses (TiMoS), with 5623 movie synopses and 95 different tropes collecting from a Wikipedia-style database, TVTropes. We present a multi-stream comprehension network (MulCom) leveraging multi-level attention of words, sentences, and role relations. Experimental result demonstrates that modern models including BERT contextual embedding, movie tag prediction systems, and relational networks, perform at most 37% of human performance (23.97/64.87) in terms of F1 score. Our MulCom outperforms all modern baselines, by 1.5 to 5.0 F1 score and 1.5 to 3.0 mean of average precision (mAP) score. We also provide a detailed analysis and human evaluation to pave ways for future research.

We study a novel task, Video Question-Answer Generation (VQAG), for challenging Video Question Answering (Video QA) task in multimedia. Due to expensive data annotation costs, many widely used, large-scale Video QA datasets such as Video-QA, MSVD-QA and MSRVTT-QA are automatically annotated using Caption Question Generation (CapQG) which inputs captions instead of the video itself. As captions neither fully represent a video, nor are they always practically available, it is crucial to generate question-answer pairs based on a video via Video Question-Answer Generation (VQAG). Existing video-to-text (V2T) approaches, despite taking a video as the input, only generate a question alone. In this work, we propose a novel model Generator-Pretester Network that focuses on two components: (1) The Joint Question-Answer Generator (JQAG) which generates a question with its corresponding answer to allow Video Question “Answering” training. (2) The Pretester (PT) verifies a generated question by trying to answer it and checks the pretested answer with both the model’s proposed answer and the ground truth answer. We evaluate our system with the only two available large-scale human-annotated Video QA datasets and achieves state-of-the-art question generation performances. Furthermore, using our generated QA pairs only on the Video QA task, we can surpass some supervised baselines. As a pre-training strategy, we outperform both CapQG and transfer learning approaches when employing semi-supervised (20%) or fully supervised learning with annotated data. These experimental results suggest the novel perspectives for Video QA training.

We introduce a novel task, Video Question Generation (Video QG). A Video QG model automatically generates questions given a video clip and its corresponding dialogues. Video QG requires a range of skills - sentence comprehension, temporal relation, the interplay between vision and language, and the ability to ask meaningful questions. To address this, we propose a novel semantic rich cross-modal self-attention (SR-CMSA) network to aggregate the multi-modal and diverse features. To be more precise, we enhance the video frames semantic by integrating the object-level information, and we jointly consider the cross-modal attention for the video question generation task. Excitingly, our proposed model remarkably improves the baseline from 7.58 to 14.48 in the BLEU-4 score on the TVQA dataset. Most of all, we arguably pave a novel path toward understanding the challenging video input and we provide detailed analysis in terms of diversity, which ushers the avenues for future investigations.