arXiv subjects

Teng Wang

Publications and source records attributed to Teng Wang.

At least 91 records · Page 5Linked to original sources

arXiv2025

Sample then Identify: A General Framework for Risk Control and Assessment in Multimodal Large Language Models

Multimodal Large Language Models (MLLMs) exhibit promising advancements across various tasks, yet they still encounter significant trustworthiness issues. Prior studies apply Split Conformal Prediction (SCP) in language modeling to construct prediction sets with statistical guarantees. However, these methods typically rely on internal model logits or are restricted to multiple-choice settings, which hampers their generalizability and adaptability in dynamic, open-ended environments. In this paper, we introduce TRON, a two-step framework for risk control and assessment, applicable to any MLLM that supports sampling in both open-ended and closed-ended scenarios. TRON comprises two main components: (1) a novel conformal score to sample response sets of minimum size, and (2) a nonconformity score to identify high-quality responses based on self-consistency theory, controlling the error rates by two specific risk levels. Furthermore, we investigate semantic redundancy in prediction sets within open-ended contexts for the first time, leading to a promising evaluation metric for MLLMs based on average set size. Our comprehensive experiments across four Video Question-Answering (VideoQA) datasets utilizing eight MLLMs show that TRON achieves desired error rates bounded by two user-specified risk levels. Additionally, deduplicated prediction sets maintain adaptiveness while being more efficient and stable for risk assessment under different risk levels.

cs.CL↗

arXiv2025

Decoupled Seg Tokens Make Stronger Reasoning Video Segmenter and Grounder

Existing video segmenter and grounder approaches, exemplified by Sa2VA, directly fuse features within segmentation models. This often results in an undesirable entanglement of dynamic visual information and static semantics, thereby degrading segmentation accuracy. To systematically mitigate this issue, we propose DeSa2VA, a decoupling-enhanced prompting scheme integrating text pre-training and a linear decoupling module to address the information processing limitations inherent in SAM-2. Specifically, first, we devise a pre-training paradigm that converts textual ground-truth labels into point-level prompts while generating corresponding text masks. These masks are refined through a hybrid loss function to strengthen the model's semantic grounding capabilities. Next, we employ linear projection to disentangle hidden states that generated by a large language model into distinct textual and visual feature subspaces. Finally, a dynamic mask fusion strategy synergistically combines these decoupled features through triple supervision from predicted text/visual masks and ground-truth annotations. Extensive experiments demonstrate state-of-the-art performance across diverse tasks, including image segmentation, image question answering, video segmentation, and video question answering. Our codes are available at https://github.com/longmalongma/DeSa2VA.

cs.CV↗

arXiv2025

Basal layer of granular flow down smooth and rough inclines: kinematics, slip laws and rheology

Granular flow down an inclined plane is ubiquitous in geophysical and industrial applications. On rough inclines, the flow exhibits Bagnold's velocity profile and follows the so-called $μ(I)$ local rheology. On insufficiently rough or smooth inclines, however, velocity slip occurs at the bottom and a basal layer with strong agitation emerges below the bulk, which is not predicted by the local rheology. Here, we use discrete element method simulations to study detailed dynamics of the basal layer in granular flows down both smooth and rough inclines. We control the roughness via a dimensionless parameter, $R_a$, varied systematically from 0 (flat, frictional plane) to near 1 (very rough plane). Three flow regimes are identified: a slip regime ($R_a \lesssim 0.45$) where a dilated basal layer appears, a no-slip regime ($R_a \gtrsim 0.6$) and an intermediate transition regime. In the slip regime, the kinematics profiles (velocity, shear rate and granular temperature) of the basal layer strongly deviate from Bagnold's profiles. General basal slip laws are developed which express the slip velocity as a function of the local shear rate (or granular temperature), base roughness and slope angle. Moreover, the basal layer thickness is insensitive to flow conditions but depends somewhat on the inter-particle coefficient of restitution. Finally, we show that the rheological properties of the basal layer do not follow the $μ(I)$ rheology, but are captured by Bagnold's stress scaling and an extended kinetic theory for granular flows. Our findings can help develop more predictive granular flow models in the future.

physics.flu-dyn↗

arXiv2025

Reinforcing Video Reasoning with Focused Thinking

Recent advancements in reinforcement learning, particularly through Group Relative Policy Optimization (GRPO), have significantly improved multimodal large language models for complex reasoning tasks. However, two critical limitations persist: 1) they often produce unfocused, verbose reasoning chains that obscure salient spatiotemporal cues and 2) binary rewarding fails to account for partially correct answers, resulting in high reward variance and inefficient learning. In this paper, we propose TW-GRPO, a novel framework that enhances visual reasoning with focused thinking and dense reward granularity. Specifically, we employs a token weighting mechanism that prioritizes tokens with high informational density (estimated by intra-group information entropy), suppressing redundant tokens like generic reasoning prefixes. Furthermore, we reformulate RL training by shifting from single-choice to multi-choice QA tasks, where soft rewards enable finer-grained gradient estimation by distinguishing partial correctness. Additionally, we propose question-answer inversion, a data augmentation strategy to generate diverse multi-choice samples from existing benchmarks. Experiments demonstrate state-of-the-art performance on several video reasoning and general understanding benchmarks. Notably, TW-GRPO achieves 50.4\% accuracy on CLEVRER (18.8\% improvement over Video-R1) and 65.8\% on MMVU. Our codes are available at \href{https://github.com/longmalongma/TW-GRPO}.

cs.CV↗

arXiv2025

EDCFlow: Exploring Temporally Dense Difference Maps for Event-based Optical Flow Estimation

Recent learning-based methods for event-based optical flow estimation utilize cost volumes for pixel matching but suffer from redundant computations and limited scalability to higher resolutions for flow refinement. In this work, we take advantage of the complementarity between temporally dense feature differences of adjacent event frames and cost volume and present a lightweight event-based optical flow network (EDCFlow) to achieve high-quality flow estimation at a higher resolution. Specifically, an attention-based multi-scale temporal feature difference layer is developed to capture diverse motion patterns at high resolution in a computation-efficient manner. An adaptive fusion of high-resolution difference motion features and low-resolution correlation motion features is performed to enhance motion representation and model generalization. Notably, EDCFlow can serve as a plug-and-play refinement module for RAFT-like event-based methods to enhance flow details. Extensive experiments demonstrate that EDCFlow achieves better performance with lower complexity compared to existing methods, offering superior generalization.

cs.CV↗

arXiv2025

Video-Holmes: Can MLLM Think Like Holmes for Complex Video Reasoning?

Recent advances in CoT reasoning and RL post-training have been reported to enhance video reasoning capabilities of MLLMs. This progress naturally raises a question: can these models perform complex video reasoning in a manner comparable to human experts? However, existing video benchmarks primarily evaluate visual perception and grounding abilities, with questions that can be answered based on explicit prompts or isolated visual cues. Such benchmarks do not fully capture the intricacies of real-world reasoning, where humans must actively search for, integrate, and analyze multiple clues before reaching a conclusion. To address this issue, we present Video-Holmes, a benchmark inspired by the reasoning process of Sherlock Holmes, designed to evaluate the complex video reasoning capabilities of MLLMs. Video-Holmes consists of 1,837 questions derived from 270 manually annotated suspense short films, which spans seven carefully designed tasks. Each task is constructed by first identifying key events and causal relationships within films, and then designing questions that require models to actively locate and connect multiple relevant visual clues scattered across different video segments. Our comprehensive evaluation of state-of-the-art MLLMs reveals that, while these models generally excel at visual perception, they encounter substantial difficulties with integrating information and often miss critical clues. For example, the best-performing model, Gemini-2.5-Pro, achieves an accuracy of only 45%, with most models scoring below 40%. We aim that Video-Holmes can serve as a "Holmes-test" for multimodal reasoning, motivating models to reason more like humans and emphasizing the ongoing challenges in this field. The benchmark is released in https://github.com/TencentARC/Video-Holmes.

cs.CV↗

arXiv2025

BPP-Search: Enhancing Tree of Thought Reasoning for Mathematical Modeling Problem Solving

LLMs exhibit advanced reasoning capabilities, offering the potential to transform natural language questions into mathematical models. However, existing open-source datasets in operations research domain lack detailed annotations of the modeling process, such as variable definitions, focusing solely on objective values, which hinders reinforcement learning applications. To address this, we release the StructuredOR dataset, annotated with comprehensive labels that capture the complete mathematical modeling process. We further propose BPP-Search, an algorithm that integrates reinforcement learning into a tree-of-thought structure using Beam search, a Process reward model, and a pairwise Preference algorithm. This approach enables efficient exploration of tree structures, avoiding exhaustive search while improving accuracy. Extensive experiments on StructuredOR, NL4OPT, and MAMO-ComplexLP datasets show that BPP-Search significantly outperforms state-of-the-art methods. In tree-based reasoning, BPP-Search excels in accuracy and efficiency, enabling faster retrieval of correct solutions. The StructuredOR dataset is available on Huggingface https://huggingface.co/datasets/LLM4OR/StructuredOR and GitHub https://github.com/LLM4OR/StructuredOR.

cs.AI↗

arXiv2025

CP-Router: An Uncertainty-Aware Router Between LLM and LRM

Recent advances in Large Reasoning Models (LRMs) have significantly improved long-chain reasoning capabilities over Large Language Models (LLMs). However, LRMs often produce unnecessarily lengthy outputs even for simple queries, leading to inefficiencies or even accuracy degradation compared to LLMs. To overcome this, we propose CP-Router, a training-free and model-agnostic routing framework that dynamically selects between an LLM and an LRM, demonstrated with multiple-choice question answering (MCQA) prompts. The routing decision is guided by the prediction uncertainty estimates derived via Conformal Prediction (CP), which provides rigorous coverage guarantees. To further refine the uncertainty differentiation across inputs, we introduce Full and Binary Entropy (FBE), a novel entropy-based criterion that adaptively selects the appropriate CP threshold. Experiments across diverse MCQA benchmarks, including mathematics, logical reasoning, and Chinese chemistry, demonstrate that CP-Router efficiently reduces token usage while maintaining or even improving accuracy compared to using LRM alone. We also extend CP-Router to diverse model pairings and open-ended QA, where it continues to demonstrate strong performance, validating its generality and robustness.

cs.CL↗

arXiv2025

Perturbative quantum Monte Carlo calculation with high-fidelity nuclear forces

Quantum Monte Carlo (QMC) is a family of powerful tools for addressing quantum many-body problems. However, its applications are often plagued by the fermionic sign problem. A promising strategy is to simulate an interaction without sign problem as the zeroth order and treat the other pieces as perturbations. According to this scheme, we construct precision nuclear chiral forces on the lattice and make perturbative calculations around a sign-problem-free interaction respecting the Wigner-SU(4) symmetry. We employ the recently developed perturbative QMC (ptQMC) method to calculate the perturbative energies up to the second order. This work presents the first ptQMC calculations for two-body next-to-next-to-next-to leading order (N$^3$LO) chiral forces and elucidates how the hierarchical nature of the chiral interactions helps organize and simplify the ptQMC calculations. We benchmark the algorithm for the deuteron, where exact solutions serve as rigorous reference points. We also reproduce the famous Tjon line by correlating the perturbative $^{4}$He binding energies with the non-perturbative $^{3}$H binding energies. These comprehensive demonstrations underscore the efficacy of ptQMC in resolving high-fidelity nuclear interactions, establishing its potential as a robust tool for \textit{ab initio} nuclear structure studies.

nucl-th↗

arXiv2025

Large Language Models are Good Multi-lingual Learners : When LLMs Meet Cross-lingual Prompts

With the advent of Large Language Models (LLMs), generating rule-based data for real-world applications has become more accessible. Due to the inherent ambiguity of natural language and the complexity of rule sets, especially in long contexts, LLMs often struggle to follow all specified rules, frequently omitting at least one. To enhance the reasoning and understanding of LLMs on long and complex contexts, we propose a novel prompting strategy Multi-Lingual Prompt, namely MLPrompt, which automatically translates the error-prone rule that an LLM struggles to follow into another language, thus drawing greater attention to it. Experimental results on public datasets across various tasks have shown MLPrompt can outperform state-of-the-art prompting methods such as Chain of Thought, Tree of Thought, and Self-Consistency. Additionally, we introduce a framework integrating MLPrompt with an auto-checking mechanism for structured data generation, with a specific case study in text-to-MIP instances. Further, we extend the proposed framework for text-to-SQL to demonstrate its generation ability towards structured data synthesis.

cs.CL↗

arXiv2025

Who is in Charge here? Understanding How Runtime Configuration Affects Software along with Variables&Constants

Runtime misconfiguration can lead to software performance degradation and even cause failure. Developers typically perform sanity checks during the configuration parsing stage to prevent invalid parameter values. However, we discovered that even valid values that pass these checks can also lead to unexpected severe consequences. Our study reveals the underlying reason: the value of runtime configuration parameters may interact with other constants and variables when propagated and used, altering its original effect on software behavior. Consequently, parameter values may no longer be valid when encountering complex runtime environments and workloads. Therefore, it is extremely challenging for users to properly configure the software before it starts running. This paper presents the first comprehensive and in-depth study (to the best of our knowledge) on how configuration affects software at runtime through the interaction with constants, and variables (PCV Interaction). Parameter values represent user intentions, constants embody developer knowledge, and variables are typically defined by the runtime environment and workload. This interaction essentially illustrates how different roles jointly determine software behavior. In this regard, we studied 705 configuration parameters from 10 large-scale software systems. We reveal that a large portion of configuration parameters interact with constants/variables after parsing. We analyzed the interaction patterns and their effects on software runtime behavior. Furthermore, we highlighted the risks of PCV interaction and identified potential issues behind specific interaction patterns. Our findings expose the "double edge" of PCV interaction, providing new insights and motivating the development of new automated techniques to help users configure software appropriately and assist developers in designing better configurations.

cs.SE↗

arXiv2025

LongVALE: Vision-Audio-Language-Event Benchmark Towards Time-Aware Omni-Modal Perception of Long Videos

Despite impressive advancements in video understanding, most efforts remain limited to coarse-grained or visual-only video tasks. However, real-world videos encompass omni-modal information (vision, audio, and speech) with a series of events forming a cohesive storyline. The lack of multi-modal video data with fine-grained event annotations and the high cost of manual labeling are major obstacles to comprehensive omni-modality video perception. To address this gap, we propose an automatic pipeline consisting of high-quality multi-modal video filtering, semantically coherent omni-modal event boundary detection, and cross-modal correlation-aware event captioning. In this way, we present LongVALE, the first-ever Vision-Audio-Language Event understanding benchmark comprising 105K omni-modal events with precise temporal boundaries and detailed relation-aware captions within 8.4K high-quality long videos. Further, we build a baseline that leverages LongVALE to enable video large language models (LLMs) for omni-modality fine-grained temporal video understanding for the first time. Extensive experiments demonstrate the effectiveness and great potential of LongVALE in advancing comprehensive multi-modal video understanding.

cs.CV↗

arXiv2025

Superconductivity above 30 K due to the introduction of oxygen in CaFeAsF

Exploring new unconventional superconductors is of great value for both fundamental research and practical applications. It is a long-term challenge to develop and study more hole-doped superconductors in 1111 system of iron-based superconductors. Here we report the discovery of superconductivity with a critical transition temperature up to 30.7 K in the compound CaFeAsF by a post-annealing treatment in air atmosphere. The superconducting behaviors are verified in both the single-crystalline and polycrystalline samples by the resistance and magnetization measurements. The analysis by combining the depth-resolved time-of-flight secondary ion mass spectrometry (TOF-SIMS) and X-ray photoelectron spectroscopy (XPS) measurements shows that the introduction of oxygen elements and the consequent changing in Fe valence by the annealing treatment may lead to the hole-type doping, which is the origin for the occurrence of superconductivity. Our results pave the way for further in-depth investigations on the hole-doped 1111 system in iron-based superconductors.

cond-mat.supr-con↗

arXiv2025

Lattice study of $J/ψ\rightarrow γη_c$ using a method without momentum extrapolation

We present a model-independent method to calculate the radiative transition without the momentum extrapolation for the off-shell transition factors. The on-shell transition factor is directly obtained from the lattice hadronic function. We apply the method to calculate the charmonium radiative transition $J/ψ\rightarrow γη_c$. After a continuous extrapolation under three lattice spacings, we obtain the on-shell transition factor as $V(0)=1.90(4)$, where the error is the statistical error that already takes into account the $a^2$-error in the continuous extrapolation. Finally, we determine the branching fraction of $J/ψ\rightarrow γη_c$ as $\operatorname{Br}(J/ψ\rightarrow γη_c)=2.49(11)_{\textrm{lat}}(5)_{\textrm{exp}}$, where the second error comes from the uncertainty of $J/ψ$ total decay width $92.6(1.7)$ keV.

hep-lat↗

arXiv2024

Tunable acoustic energy concentrations based on pseudo-spin locking waveguides and topological rainbow trappings

In this work, tunable acoustic energy concentrations are realized based on pseudo-spin locking waveguides and topological rainbow trappings. Firstly, a tunable pseudo-spin locking is proposed, and the broad acoustic energy transport and spin-locked one-way transport are verified. The results show that acoustic wave transports based on pseudo-spin locking waveguides are more robust to structure defects than conventional topological edge-state waveguides. Besides, topological rainbow trappings are realized by adjusting the distribution of liquid in tubes. Based on those, we investigate the coupling of the pseudo-spin locking waveguides and the topological rainbow trappings. The results show that high acoustic energy concentrations can be obtained conveniently by using the coupling energy concentrator based on the pseudo-spin locking waveguides and the topological rainbow trappings. The results present a novel method to concentrate acoustic wave energy, which is vital in the application field of acoustic sensings and microfludics. The according experimental verifications will be presented in the near future.

physics.app-ph↗

arXiv2024

Auto-ABSA: Cross-Domain Aspect Detection and Sentiment Analysis Using Auxiliary Sentences

After transformer is proposed, lots of pre-trained language models have been come up with and sentiment analysis (SA) task has been improved. In this paper, we proposed a method that uses an auxiliary sentence about aspects that the sentence contains to help sentiment prediction. The first is aspect detection, which uses a multi-aspects detection model to predict all aspects that the sentence has. Combining the predicted aspects and the original sentence as Sentiment Analysis (SA) model's input. The second is to do out-of-domain aspect-based sentiment analysis(ABSA), train sentiment classification model with one kind of dataset and validate it with another kind of dataset. Finally, we created two baselines, they use no aspect and all aspects as sentiment classification model's input, respectively. Compare two baselines performance to our method, found that our method really makes sense.

cs.CL↗

arXiv2024

ImagineNav: Prompting Vision-Language Models as Embodied Navigator through Scene Imagination

Visual navigation is an essential skill for home-assistance robots, providing the object-searching ability to accomplish long-horizon daily tasks. Many recent approaches use Large Language Models (LLMs) for commonsense inference to improve exploration efficiency. However, the planning process of LLMs is limited within texts and it is difficult to represent the spatial occupancy and geometry layout only by texts. Both are important for making rational navigation decisions. In this work, we seek to unleash the spatial perception and planning ability of Vision-Language Models (VLMs), and explore whether the VLM, with only on-board camera captured RGB/RGB-D stream inputs, can efficiently finish the visual navigation tasks in a mapless manner. We achieve this by developing the imagination-powered navigation framework ImagineNav, which imagines the future observation images at valuable robot views and translates the complex navigation planning process into a rather simple best-view image selection problem for VLM. To generate appropriate candidate robot views for imagination, we introduce the Where2Imagine module, which is distilled to align with human navigation habits. Finally, to reach the VLM preferred views, an off-the-shelf point-goal navigation policy is utilized. Empirical experiments on the challenging open-vocabulary object navigation benchmarks demonstrates the superiority of our proposed system.

cs.RO↗

arXiv2024

Leveraging Large Language Models for Solving Rare MIP Challenges

Mixed Integer Programming (MIP) has been extensively applied in areas requiring mathematical solvers to address complex instances within tight time constraints. However, as the problem scale increases, the complexity of model formulation and finding feasible solutions escalates significantly. In contrast, the model-building cost for end-to-end models, such as large language models (LLMs), remains largely unaffected by problem scale due to their pattern recognition capabilities. While LLMs, like GPT-4, without fine-tuning, can handle some traditional medium-scale MIP problems, they struggle with uncommon or highly specialized MIP scenarios. Fine-tuning LLMs can yield some feasible solutions for medium-scale MIP instances, but these models typically fail to explore diverse solutions when constrained by a low and constant temperature, limiting their performance. In this paper, we propose and evaluate a recursively dynamic temperature method integrated with a chain-of-thought approach. Our findings show that starting with a high temperature and gradually lowering it leads to better feasible solutions compared to other dynamic temperature strategies. Additionally, by comparing results generated by the LLM with those from Gurobi, we demonstrate that the LLM can produce solutions that complement traditional solvers by accelerating the pruning process and improving overall efficiency.

cs.CL↗