Artificial Intelligence · Cross-listed from cs.CL
Diagnosing the Fact-Grounding Gap in Multi-Hop Question Answering
Kevin Mo, Nathan Mo and Richard Zhu
Multi-hop question answering requires combining information from multiple documents to answer complex questions. These systems have grown increasingly capable, yet when they fail, the error is typically attributed to not finding the right documents. Whether this holds at the level of individual
reasoning steps remains largely unexamined. We investigate this across three standard multi-hop QA
benchmarks and find that failures decompose into two distinct modes:
retrieval failures, where the needed passage was not retrieved, and extraction failures, where the passage was retrieved but the needed fact could not be extracted - a phenomenon we term the fact-grounding gap. Extraction failures account for nearly half of all per-hop deficiencies and are invisible to standard
retrieval metrics. They remain unresolved by every
retrieval intervention we test, establishing a ceiling for
retrieval-only improvements. The gap's severity varies across
benchmarks and question types, but extraction failures appear on every
dataset we measure. Our findings reveal that
retrieval failures and extraction failures are fundamentally different bottlenecks requiring different solutions - a distinction absent from current evaluation practice.
Read the full abstract ▾
Artificial Intelligence
Sparse MLLM Anchors, Dense Adaptation: Breaking the Self-Referential Loop in Wild Test-Time Adaptation
Zhenbin Wang, Lei Zhang, Lituan Wang et al. (+3 more)
Wild test-time adaptation (WTTA) updates a source model online under small test batches, concurrent distribution shifts, and time-varying class imbalance. Most WTTA methods derive their adaptation signals, including predictive uncertainty, sample reliability, and local feature geometry, from the model being adapted. When the source model is unreliable under shift, these signals can reinforce its own errors, forming a self-referential loop. We introduce MASA (
Multimodal-LLM-Anchored Semantic Adaptation), which complements model-internal evidence with structured semantic descriptions from a frozen
multimodal large
language model (MLLM). To limit inference cost, MASA queries the MLLM only for a small set of diverse, reliability-ranked anchors. The resulting descriptions capture the object family and nuisance factors such as style, viewpoint, and occlusion. MASA encodes these descriptions, propagates them to neighboring test samples, and stores the resulting visual-semantic information in an online prototype
memory. Descriptor-aware
retrieval from this
memory provides an auxiliary target for lightweight adaptation of normalization-affine parameters. We evaluate MASA on the WTTA ImageNet-C
benchmark under limited-batch, mixed-domain, and imbalanced-label-shift settings with ResNet and ViT backbones.
Read the full abstract ▾
All authors Zhenbin Wang, Lei Zhang, Lituan Wang, Yan Wang, Zhao Zhang, Wei Huang
Artificial Intelligence · Cross-listed from cs.LG
Distributed JEPA: A Self-Supervised Framework for Energy Forecasting
Liana Toderean, Tudor Cioara, Vasilis Michalakopoulos et al. (+3 more)
Traditional energy
forecasting solutions rely on task-specific supervision and energy asset representations, limiting transferability and the ability to capture general temporal dynamics across heterogeneous assets. We address this by proposing a distributed Joint
Embedding Predictive Architecture (JEPA) for self-supervised learning from heterogeneous energy time-series. The framework predicts latent representations of masked temporal segments while integrating temporal observations and contextual information within a shared
embedding space. To prevent representation collapse, training combines a latent-space predictive objective with covariance and temporal variance regularization. The evaluation was conducted on energy consumption and generation
datasets under data-degradation scenarios and compared with a
Transformer forecasting baseline. The learned representations remained stable (cosine similarity $\approx 0.98$; effective rank 185-235). JEPA achieved performance comparable to a
Transformer on building energy data, higher $R^2$ in 3/5 consumer clusters, and outperformed the baseline on 9/10 unseen PVs ($R^2$=0.73-0.88 vs. <0.45), while showing greater robustness to missing data.
Read the full abstract ▾
All authors Liana Toderean, Tudor Cioara, Vasilis Michalakopoulos, Efstathios Sarantinopoulos, Ionut Anghel, Elissaios Sarmas
Artificial Intelligence
SKIP: a Self-knowledge-guided Step-wise Preference Learning Framework for Concise Reasoning
Qinhong Lin, Yuhao Zhang, Yinglun Feng et al. (+2 more)
While Chain-of-Thought (CoT)
reasoning has been proven to be effective, it often leads to overthinking, resulting in computational overhead, inference latency, and even degraded performance in large
language models (LLMs). Existing concise
reasoning frameworks significantly compromise accuracy while compressing the length of output. In this paper, we propose SKIP, a self-knowledge-guided step-wise preference learning framework. Starting with lightweight fine-tuning to adjust the model's output style, SKIP introduces a carefully designed knowledge probing mechanism to guide model to output an answer at each
reasoning step. Based on the correctness of intermediate steps, we construct preference data that guide the model toward more efficient and correct
reasoning by leveraging DPO. Experimental results demonstrate that our method effectively improves
reasoning compression while mitigating performance degradation after fine-tuning. Besides, SKIP shows strong generalization ability on out-of-distribution
datasets. We further conducted ablation studies on the component parameters of our framework.
Read the full abstract ▾
All authors Qinhong Lin, Yuhao Zhang, Yinglun Feng, Zhongliang Yang, Linna Zhou
Artificial Intelligence
ORDER: Task-Conditioned Routing for Retrieval-Augmented Generation
Aurélien Pellet, Julien Perez and Marie Puren
Retrieval-Augmented Generation (RAG) pipelines typically rely on a fixed indexing and
retrieval configuration determined at preprocessing time. This one-size-fits-all design is ill-suited to domain-expert settings, where heterogeneous queries require different chunking granularities, metadata constraints, and source-selection strategies. As a result, configurations that are effective for one family of queries often perform poorly for others. In this paper, we introduce ORDER (Optimal Routing for Dynamic Evidence
Retrieval), a query-conditioned RAG framework that jointly adapts indexing and
retrieval to the incoming query. Our approach first discovers semantic clusters over a given set of questions associated to a corpus and learns, for each cluster, a chunking strategy together with a suited metadata filtering and reranking configuration. At inference time, queries are routed to the appropriate pre-built index through nearest-centroid assignment. To further improve
retrieval, we propose a supervised query router (QRe) that predicts which collections are most likely to contain relevant evidence, coupled with a Uniform Multi-source Sampler (UMS) that allocates the
retrieval budget evenly across the selected sources. We evaluate our framework on large-scale, heterogeneous historical archives and show that conditioning both indexing and
retrieval on the query consistently outperforms both naive baselines and strong state-of-the-art RAG systems in complex expert-domain environments.
Read the full abstract ▾
Artificial Intelligence
ThinkFlow: Self-Evolving Probabilistic Latent Memory for Lifelong Conversational Agents
Cai Ke, Xin Liu, Han Zhang et al. (+6 more)
Lifelong conversational
agents rely on
memory systems to maintain deep, context-aware interactions with users. However, existing explicit textual
memory pipelines suffer from a severe information bottleneck, often losing subtle behavioral patterns and emotional shifts. Furthermore, being typically static post-deployment, they cannot autonomously adapt to personal habits and preferences without manual feedback. Cognitive science, however, suggests that humans maintain mental models purely in a latent space and continuously refine them through predictive coding. Inspired by this, we propose ThinkFlow, a novel end-to-end latent
memory framework for lifelong conversational
agents. ThinkFlow bypasses the text bottleneck by dynamically compressing conversational flows into probabilistic latent
memory skills, autonomously consolidating complex user states into disentangled, continuous vectors without semantic interference. To break this barrier, we introduce a test-time
evolution paradigm. By coupling teacher-guided latent
alignment to bootstrap the initial state with a self-supervised next-user-utterance prediction task for continuous refinement, the framework successfully overcomes cold-start challenges and achieves label-free lifelong personalization. Extensive experiments on long-term conversation
benchmarks demonstrate that ThinkFlow significantly outperforms prevailing
memory systems, providing highly personalized and contextually accurate responses over extended multi-session interactions.
Read the full abstract ▾
All authors Cai Ke, Xin Liu, Han Zhang, Jiangyue Yan, Zike Yuan, Ling Deng, Yue Yu, Hui Wang, Ruifeng Xu
Artificial Intelligence
FlexEE: Self-Speculative and KV-Compatible Early Exiting for Offloading-Aware LLM Inference
Qihu Xie, Ziwei Li and Yi Kang
Large
language model (LLM) inference is often constrained by both computation and
memory, especially in offloading-based deployments where model weights are transferred across
memory hierarchies during autoregressive decoding. In this setting, reducing the number of executed layers can lower per-
token latency while also avoiding costly weight movement. Motivated by this observation, we present FlexEE, an early exiting framework for resource-constrained and offloading-based LLM inference. FlexEE makes early exiting practical for LLM decoding through layer-wise exit supervision for reliable intermediate-layer prediction, self-speculative decoding over a Top-K local vocabulary for low-cost exit decisions, and dynamic hidden state management for KV-cache-correct and
memory-aware execution. Across generative and downstream tasks, FlexEE enables efficient early exit with minimal accuracy degradation, delivering up to 1.27$\times$/3.16$\times$ and 1.25$\times$/2.83$\times$ end-to-end speedups on Llama2-7B and Llama3-8B under 0%/50% weight offloading, respectively.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CL
The Role of Implicit and Explicit Demographic Signals in Large Language Model-based Student Assessment
Donya Rooein, Luca Benedetto and Dirk Hovy
Large
Language Models are now common in student assessment, but we know little about how student demographics affect their use. Sometimes, considering student demographics may be necessary – for example, to improve readability for users with lower educational levels. However, it also risks being a cause of discrimination, e.g., when assigning lower scores to students from lower socioeconomic backgrounds. We set up controlled
prompts to test 1) explicit demographic effects, where we mention demographic details directly, and 2) implicit effects, where we use conversation history as a demographic signal. We test these settings in three tasks: Automated Essay Scoring, Formative Feedback, and Metalinguistic Question Answering. We test six state-of-the-art LLMs on these tasks. In both explicit and implicit cases, the models pick up on demographic cues and can change their scoring, feedback, and answers accordingly. We find that LLMs frequently adjust the readability of feedback to education levels when these are explicitly mentioned. On the other hand, implicit conditions produce unpredictable biases, such as in question answering, where responses from lower-education levels receive lower sentiment scores. Our results provide clear evidence of demographic sensitivity in LLMs for educational assessment tasks.
Read the full abstract ▾
Artificial Intelligence
Affect-Prototype Guided Fusion for Open-Vocabulary Incomplete Multi-modal Emotion Recognition
Yichi Zhang, Shenyue Wang, Jing Luo et al. (+2 more)
Open-vocabulary
multimodal emotion recognition (OV-MER) aims to generate open natural-
language emotion labels from
multimodal affective cues. In real-world scenarios, however, complete and synchronized modal data are difficult to obtain due to limitations of acquisition devices and user
privacy constraints. Existing OV-MER methods are largely designed for full-modal inputs, and fail to perform effective feature fusion under modal missing conditions. Meanwhile, current fusion approaches designed for incomplete modalities mainly focus on fixed-label recognition context, and cannot satisfy the demand for fuse emotional cues guided with arbitrary
emotion semantics in OV-MER context. To tackle these challenges, this paper proposes an Affect-Prototype-Conditioned Fusion (APCF) framework for incomplete open-vocabulary
emotion recognition. As a candidate-free generative framework, APCF extends modal contribution learning to scenarios guided by arbitrary emotional semantics. Specifically, we construct an affect-prototype library to explicitly model
multimodal contribution characteristics corresponding to diverse
emotions, which provides dynamic constraints for modal fusion under different emotional semantic perspectives. Conditional
retrieval and feature aggregation are conducted based on available modal features. The refined fused affective representations are then fed into an LLM decoder to produce open-vocabulary
emotion labels. Experiments on the OV-MERD+ and MER-FG
datasets demonstrate that APCF substantially outperforms state-of-the-art baselines.
Read the full abstract ▾
All authors Yichi Zhang, Shenyue Wang, Jing Luo, Chunyang Yu, Xinyu Yang
Artificial Intelligence
AntennaFlow: A Generative Flow Model for Offset Correction in Phaseless Antenna Testing
Yongzhi Li, Chongting Shen, Menglin Chen et al. (+2 more)
Near-field to far-field transformation is central to large-aperture
antenna testing, yet two coupled challenges remain: costly phase acquisition at millimeter-wave bands and violations of the centering assumption under offset mounting. Existing methods address these issues separately, requiring either dense full-field data or offset vectors. We tackle both jointly by exploiting a key observation: amplitude fields under different offsets are coordinate-transformed views of the same near field. The challenge is to recover the center-aligned field from offset amplitudes without a phase or offset vector. We propose AntennaFlow, a three-stage framework: a contrastively learned encoder that maps offset views to an offset-invariant
embedding, a deterministic flow-matching transport that maps offset amplitudes to center-aligned ones, and the Simplified Extrapolation Technique, whose Green-function Taylor expansion is valid only for centered fields. Experiments show that AntennaFlow enables fast, phaseless, offset-vector-free NF–FF reconstruction from sparse amplitude-only measurements, consistently outperforming existing baselines while preserving physical consistency.
Read the full abstract ▾
All authors Yongzhi Li, Chongting Shen, Menglin Chen, Xun Jiang, Zhengpeng Wang
Artificial Intelligence · Cross-listed from cs.NI
AeroLat: Channel-Aware Latent Space Semantic Communication for Decentralized UAV Swarms
Rajdeep Ghosh, Goparaju Venkata Seshachala Sree Vatsava and Sudip Misra
Communication in latent space offers an intriguing alternative to symbolic messages for decentralized autonomous Unmanned Aerial Vehicle (
UAV) swarms operating over bandwidth-constrained, time-varying wireless links. However, when homogeneous frozen models are prompted with discretized perceptual inputs, their broadcast states collapse toward the shared
prompt template. In view of this, we propose AeroLat, a channel-aware latent semantic communication framework that uses evidence injection. The resulting latent states are then passed through an explicit communication model that encompasses bandwidth-limited serialization, additive noise and information staleness, which facilitates a joint assessment of communication fidelity and swarm-level coordination. Across multi-seed simulations, AeroLat provably remains resilient to codec choice, faults and increasing swarm size. It consistently reproduces the latent-swarm anomaly, while no-whitening controls recover the collapse. In particular, AeroLat is capable of reducing false similarity by 97.5%.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Beyond Token-Local Imitation: Reward-Compatible Temporal Credit Assignment for On-Policy Distillation
Shiqi Liu, Zeyu He, Letian Tao et al. (+9 more)
On-policy
distillation (OPD) has emerged as an effective approach for large
language model post-training, yet existing objectives face a trade-off between objective fidelity and optimization stability.
Token-level OPD provides stable but local supervision, whereas sequence-level OPD captures future credit at the cost of horizon-dependent variance. We establish a unified temporal-credit view of these formulations, showing that practical
token-level OPD can be interpreted as a temporal approximation to the sequence-level reverse-KL gradient. Building on this connection, we propose $\gamma$OPD (GammaOPD), which uses discounted temporal credit assignment to balance long-horizon supervision and optimization stability, while admitting a horizon-independent variance bound. We further develop a reward-compatible bounded mixing (RBM) mechanism for $\gamma$OPD that balances verifiable outcome feedback with the discounted OPD advantage to move beyond purely teacher-dependent optimization. Experiments on mathematical and code
reasoning demonstrate consistent improvements over existing OPD methods across vanilla, size-mismatched, and multi-teacher
distillation settings.
Read the full abstract ▾
All authors Shiqi Liu, Zeyu He, Letian Tao, Guojian Zhan, Jiaxin Gao, Feihong Zhang, Jingliang Duan, Wei Xiong, Kehua Sheng, Bo Zhang, Yang Guan, Shengbo Eben Li
Artificial Intelligence · Cross-listed from cs.SE
RepoAtlas: Guiding Coding Agents via Evolving Multimodal Repository Views
Yunxiang Zhang, Haiquan Wang, JiaWei Guo et al. (+5 more)
Large
language model (LLM)-powered coding
agents have made rapid progress in automating software engineering tasks, yet repository-level issue resolution remains challenging. Beyond generating a plausible patch, an
agent must localize relevant code across interdependent files and maintain repository context that is both sufficient and focused. Code graphs expose non-local relations, but linear text interfaces obscure their
topology; rendering the full repository graph yields visual representations that are too dense to perceive reliably, whereas a one-shot local view becomes stale as exploration proceeds. We present RepoAtlas, a training-free module that maintains evolving
multimodal repository views through a select–project–refresh loop over a repository code graph. RepoAtlas combines evidence from the issue with the
agent's current exploration state to select a task-relevant region under a fixed budget, projects the selected structure into complementary visual and textual representations, and refreshes the view when changes in the exploration state render it outdated. We evaluate RepoAtlas on SWE-bench Verified, where it improves the resolve rate by 2.4 points while reducing input
tokens and model calls by 5.8% and 7.8% on average, relative to the strongest
multimodal graph baseline, with consistent gains across three models of different families and scales.
Read the full abstract ▾
All authors Yunxiang Zhang, Haiquan Wang, JiaWei Guo, Hanyang Xia, Yan Chen, Tong Chen, Zhang Zhiwei, Junchen Ye
Artificial Intelligence · Cross-listed from cs.LG
When Confidence Signals Disagree: Local and Global Confidence in Autoregressive Language Models
Julio C. Amador Diaz Lopez
Modern predictive systems expose multiple quantities that are commonly interpreted as measures of confidence. However, these quantities can summarize different aspects of the predictive process. This distinction matters when confidence is used to evaluate reliability or inform downstream oversight and control. We investigate whether different confidence readouts are empirically interchangeable in an autoregressive
language model by comparing local confidence, defined from the probability of the greedy-selected answer
token, with global confidence, defined from modal-answer frequency under repeated sampling. Across MMLU and ARC Challenge, the two signals are weakly correlated and differ substantially in their association with correctness: global confidence is moderately associated with correctness, whereas local confidence shows little association. We further test whether question-level disagreement between the signals is associated with sampling instability. On ARC, larger local–global confidence gaps are associated with higher answer
entropy, more distinct sampled answers, and lower modal-answer concentration. The gap–
entropy association persists when disagreement and instability are estimated from disjoint
stochastic samples, indicating that it is not explained by shared finite-sample variation. The corresponding relationship is substantially weaker on MMLU, where only 4% of questions exhibit sampling instability. These results show that confidence readouts derived from the same predictive system are not empirically interchangeable and that their disagreement can provide a diagnostic of unstable sampling behavior. Confidence should therefore be treated as an explicitly defined measurement rather than as a single intrinsic scalar property of a model, particularly when it is used to inform downstream evaluation, oversight, or control.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from stat.ME
Causal Discovery via Transformed Low-Rank Quantile Surfaces
Ryo Kamimura and Thong Pham
We propose Low-Rank
Quantile Surfaces (LRQS), a bivariate causal model in which, in the causal direction, an unknown monotone transformation of the conditional
quantile surface admits a low-rank functional decomposition. LRQS subsumes location-scale noise models and post-nonlinear heteroscedastic noise models, while allowing multiple
quantile bases to represent changes beyond location-scale effects. We prove generic identifiability of LRQS: the transformed
quantile surface is low rank in the causal direction, whereas reverse representability under the corresponding constraints occurs only for exceptional, fine-tuned cause marginals. We provide a nonparametric causal score using a fitting procedure that alternates between rank-constrained approximation of discretized
quantile surfaces and isotonic estimation of the unknown monotone transformation. Experiments on synthetic mechanisms with higher-rank distributional shape variation and strong nonlinear distortions, together with standard bivariate
benchmarks, show that LRQS is especially effective when conditional distributional shape or observation distortion goes beyond existing location-scale assumptions.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Repurposing Deep Limit Order Book Forecasting for Scenario-Conditioned Market Impact Modeling
Eljas Linna, Kestutis Baltakys, Derrick Manoharan et al. (+2 more)
Deep Limit Order Book
forecasting models capture nonlinear
market dynamics, but their ability to quantify the effects of counterfactual order book messages has not been systematically validated. We introduce a model-agnostic framework that compares a trained forecaster's predictive distributions before and after injecting mechanically valid counterfactual messages, defining short-horizon model-implied
market impact. A
Transformer-based forecaster recovered scenario rankings with a Spearman correlation of 0.99 and 97.2% directional agreement with realized historical outcomes among non-neutral scenarios. Observation-level analysis further showed that estimated impacts captured incremental sequence-dependent variation beyond scenario identity and the pre-event forecast. These results provide evidence that pretrained Limit Order Book forecasters can be repurposed for scenario-conditioned response modeling without retraining.
Read the full abstract ▾
All authors Eljas Linna, Kestutis Baltakys, Derrick Manoharan, Alexandros Iosifidis, Juho Kanniainen
Artificial Intelligence · Cross-listed from cs.LG
Verbalizing Subliminal Learning Effects Using Text Optimization
Nathan Hu, Sanmi Koyejo and Christopher Potts
Subliminal learning is a phenomenon in which a
distillation dataset transmits traits from the teacher model that are not legibly encoded in the
dataset itself. This introduces a new challenge for model development and creates new risks from data poisoning. In this work, we use text optimization to detect subliminal learning effects and describe them as legible
prompts. Subliminal learning from a prompted teacher motivates our approach. We observe that this is a special case of context
distillation and leverage this observation to show that, in theory, the prompted subliminal learning
dataset identifies the teacher's
prompt. We reduce recovering this
prompt to a text optimization problem and present a method to approximately solve it. Our method, SALVE (Search-Aided Latent Verbalization), optimizes a soft
prompt, queries the same model to verbalize it as text, and uses beam search to make the verbalization reliable. In the standard subliminal learning setting, SALVE reliably recovers legible
prompts that name the teacher's trait, while common text optimization methods fail to do so. In addition, we find that there are settings in which SALVE recovers the teacher's trait from a
dataset even when subliminal learning fails, but that modifying student training to improve context
distillation can create subliminal learning effects. We lastly show that SALVE detects subliminal learning effects in three additional settings: (1) mixtures of subliminal learning data and unrelated data, (2) data generated when the teacher is biased via activation steering, and (3) subsets of real preference data selected via Logit-Linear Selection. Overall, our results deepen our understanding of subliminal learning and present SALVE as a method to proactively detect subliminal learning effects.
Read the full abstract ▾
Artificial Intelligence
QART: A Quantum-Classical Hybrid Architecture for Long-Horizon Reasoning – Exploring a Conditional Path toward Quantum Scaling
Lehao Lin, Yuheng Cheng, Guolong Liu et al. (+9 more)
Long-horizon
reasoning is vulnerable to early errors that compromise later decisions. We present QART, the
Quantum-Augmented
Reasoning Transformer, a
quantum–classical hybrid architecture combining a backbone
language model with
quantum encoding, CIM-based QUBO optimization, and
quantum decoding. Semantic information can come from hidden representations or model-generated text; detailed encoding and optimization procedures remain proprietary. Under explicit assumptions, we establish a conditional asymptotic reliability separation from single-trajectory autoregressive LLMs. For a common task family with aligned optimality and acceptance criteria, autoregressive acceptance probability tends to zero when cumulative conditional risk of irreversible errors diverges. QART's task-optimal-path recovery probability remains bounded away from zero if conditional probabilities for optimal-path coverage and semantic fidelity, spectral certification, dynamical reachability, and faithful readout remain uniformly positive under a specified resource schedule. The architecture alone does not imply these bounds. Paired measurements on six long-horizon
benchmarks using DeepSeek V4 Flash, GLM-5.3, and GPT-5.5 xhigh in a Codex
agent environment favor QART in 14 of 15 backbone–
benchmark pairs. Relative gains reach 84.0% on SciCode, 47.6% on $\tau^3$-Bench, and 44.4% on Terminal-Bench 4.0; the DeepSeek V4 Flash configuration regresses by 7.8% on DeepSWE. These results do not directly validate the asymptotic separation. Potential
quantum scaling laws are formulated as conditional hypotheses. A
quantum-advantage interpretation requires a demonstrated CIM
quantum advantage over strong classical solvers and its transfer to end-to-end
reasoning after all system overheads.
Read the full abstract ▾
All authors Lehao Lin, Yuheng Cheng, Guolong Liu, Yao Li, Xuning Tan, Xiyuan Zhou, Ruixi Zou, Shi Wang, Huan Zhao, Wenxuan Liu, Haifeng Wu, Junhua Zhao
Artificial Intelligence
Bridging Learned Visual Perception and Symbolic Belief-Space Planning
Guy Azran, Michael Navat and Sarah Keren
In partially observable settings,
agents must act without full knowledge of the world state and rely on uncertain state-estimation pipelines. Obtaining grounded and verifiable symbolic plans under such uncertainty remains a key challenge. Recent work has integrated
Vision-
Language Models (VLMs) to bridge perception and symbolic
reasoning, following two main paradigms. The first, VLM-as-planner, maps images directly to action sequences, and the second, VLM-as-grounder, grounds observations into symbolic predicates used as the initial state by off-the-shelf planners. Both approaches ignore uncertainty in the planning process, compromising robustness. We introduce a third paradigm, VLM-as-probabilistic-grounder, a novel approach that captures the uncertainty of VLM predicate groundings as a probability distribution over symbolic states. This enables planning in belief space and producing robust plans under uncertainty. Experiments in simulated household
robot settings show improved robustness and task success over deterministic grounding, underscoring how our approach leverages foundation models for reliable planning under uncertainty.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
VOR-Bench: A Human Perception-Driven Benchmark for Video Object Removal
Haonan Huang, Tianrui Qiu, Xianghao Zang et al. (+10 more)
Despite its crucial role in
video object removal (VOR), existing evaluation paradigms face two critical limitations: questionable references and a misalignment between tradi- tional metrics and human preference. To address these challenges, we introduce VOR- Bench, which advances VOR evaluation through three integrated components. First, we present the VOR
Dataset (VORD), the first
benchmark dataset providing both paired edited
videos and graffiti masks. Its unique strength lies in a diverse data spectrum, which encompasses model-generated, tool-rendered, and camera-captured data, ensuring robust assessment across real-world scenarios. Second, we develop rMPAF, a realistic Motion- capable Paired-
video Acquisition Framework. By combining the strengths of image- based object removal and fine-tuned
video generation models, rMPAF automatically generates realistic, motion-coherent paired
videos. Finally, we propose three evaluation dimensions and introduce VOR-MDSM, the first perception-driven VLM-based scoring model specifically designed for mask-guided VOR. It bridges the gap between arithmetic metrics and human perception by covering the essential visual attributes and matching nuanced human judgment. Extensive experiments demonstrate that VOR-Bench yields evaluation results that align closely with human perception, achieving a remarkable cor- relation (h̊o} > 0.9) with subjective assessments. We will release VOR-Bench along with its documentation to ensure full reproducibility.
Read the full abstract ▾
All authors Haonan Huang, Tianrui Qiu, Xianghao Zang, Yinan Du, Zhixiang He, Chi Zhang, Hao Sun, Zhongjiang He, Tianwei Cao, Xuchong Zhang, Hongbin Sun, Kongming Liang, Zhanyu Ma
Artificial Intelligence · Cross-listed from cs.HC
The Evolution of Coordination in a Collective Intelligence System: 25 Years of English Wikipedia and the Emergence of Generative AI
Neal Reeves, Maja Świeczkowska, Amy Rechkemmer et al. (+1 more)
English Wikipedia is one of the largest examples of collective intelligence on the Web, sustained not only by article production but also by volunteer coordination and governance. While prior research has examined coordination work in Wikipedia, less
attention has been paid to how participation in these spaces has evolved over time. Drawing on a longitudinal analysis spanning nearly 25 years of English Wikipedia, we examine editing patterns across five namespaces covering content, discussion, and governance. We find that participation in coordination spaces has declined relative to content production, particularly in governance areas, with a shrinking core of editors performing an increasing share of this work. Using Markov-based session metrics, we also find that editing has become more specialised, with editors moving less frequently between namespaces. Motivated by recent governance debates around generative AI, we conclude by investigating whether the availability of LLMs has altered these long-term trends. While short-term changes are visible, we find little evidence that generative AI fundamentally changed existing trajectories of coordination and participation.
Read the full abstract ▾
All authors Neal Reeves, Maja Świeczkowska, Amy Rechkemmer, Elena Simperl
Artificial Intelligence
CoAdapt: An LLM-based Framework for Adaptive Collaborative Perception in IIoT Robotic Swarms
Houssam Hajj Hassan, Antonia Maria Masucci, Lynda Zitoune et al. (+1 more)
Industrial IoT environments increasingly deploy autonomous mobile
robots for tasks such as material handling, product assembly, or infrastructure inspection. In such deployments, collaborative perception enables
robots to share
LiDAR observations and collectively construct a richer model of their environment than an individual
agent could produce alone. However, industrial environments are dynamic spaces where
robot positions shift continuously, network bandwidth fluctuates, and the marginal contribution of
robots to perception quality varies at runtime. Existing collaborative perception approaches are designed for static participation assumptions and cannot adapt to these dynamics without sacrificing either detection precision or communication efficiency. This paper presents CoAdapt, an adaptive collaborative perception framework for IIoT robotic swarms in which a Large
Language Model (LLM) serves as a runtime fusion controller, jointly deciding which
robots participate in the fusion process and which fusion
algorithm to apply based on the current spatial configuration and network state. The LLM reasons over structured natural
language descriptions of the scene derived from raw
LiDAR point clouds, requiring no taskspecific training and generalizing to unseen swarm
topologies. Evaluated on the OPV2V
benchmark across 25 scenarios, our approach achieves a 38% reduction in communication cost while maintaining detection precision comparable to static baseline approaches.
Read the full abstract ▾
All authors Houssam Hajj Hassan, Antonia Maria Masucci, Lynda Zitoune, Salah-Eddine Elayoubi
Artificial Intelligence · Cross-listed from cs.CV
RegRet: Enhancing Region-Level Retrieval in Large Multimodal Models
Xun Liang, Honghui Yang, Weihang Pan et al. (+6 more)
Region-level
retrieval aims to align user-specified image regions with relevant regions or textual descriptions, playing a crucial role in realworld applications such as e-commerce product search and RAG. Although recent Large
Multimodal Models (LMMs) have made significant strides in
multimodal retrieval, they primarily focus on global-level tasks and struggle to capture effective region-level representations. To bridge this gap, we present RegRet, an LMM-based Region-level
Retrieval framework that enhances the regional representations without compromising overall global
retrieval performance. At its core, RegRet integrates a Region-Aware Encoder to capture detailed regional features while balancing them with the global background context. To further enhance the fine-grained understanding and discriminability of representations, we design a multi-stage training pipeline that includes detailed localized captioning and regional contrastive learning tasks. In addition, considering the absence of region-level contrastive training data and the limited diversity of evaluation tasks in current
benchmarks, we introduce the REGMB
benchmark. It comprises 225k contrastive pairs, covering four
multimodal retrieval tasks. Extensive experiments validate the effectiveness of our approach. RegRet outperforms strong baselines in the zero-shot setting. Further training with contrastive learning leads to an average improvement of more than 20% on both REGMB and public
benchmarks, while achieving comparable or better results on global-level
retrieval tasks.
Read the full abstract ▾
All authors Xun Liang, Honghui Yang, Weihang Pan, Ruisi Zhao, Boyuan Pan, Yao Hu, Wenxiao Wang, Binbin Lin, Deng Cai
Artificial Intelligence · Cross-listed from cs.CV
StackTok: Accelerating VLMs Inference with Budget-Adaptive Visual Token Selection
Zhenbin Wang, Lei Zhang, Lituan Wang et al. (+3 more)
Increasing image resolution produces ever-longer visual-
token sequences in
vision-
language models (VLMs), substantially raising their inference cost. To reduce this overhead without retraining, existing methods select compact
token subsets that prioritize query relevance, visual coverage, or a fixed trade-off between them. The appropriate balance, however, varies across queries and
token budgets: localized questions favor relevance, whereas holistic questions demand broader visual coverage. We introduce StackTok, a training-free selector that treats query relevance as the objective and visual coverage as budget-calibrated support. StackTok builds a size-indexed coverage reference from a coverage-only greedy sequence and adjusts its support target using query–
vision affinity
entropy. A reference-gated interleaved selection policy then switches between relevance- and coverage-oriented additions according to the current subset's support deficit. For high-resolution inputs, StackTok allocates one shared
token budget across crops according to the combined marginal gain of locally nominated
tokens. Evaluated with five VLMs over ten distinct image-understanding
benchmarks, StackTok ranks first among training-free selectors in every tested model–budget setting. On high-resolution LLaVA-NeXT-7B, it retains 95.26% of full-
token performance with only 160 of 2{,}880 (5.6%) visual
tokens.
Read the full abstract ▾
All authors Zhenbin Wang, Lei Zhang, Lituan Wang, Wei Huang, Yan Wang, Zhenwei Zhang
Artificial Intelligence · Cross-listed from cs.CV
What Breaks Local Watermarks? A Robustness Benchmark for Local Invisible Image Watermarking
Kai Yao, Bence Szilágyi, Sebestyén Kamp et al. (+4 more)
Local image watermarking embeds an invisible signal into selected image regions rather than spreading it across the entire image, enabling payload recovery from specific objects or regions without perceptibly altering the image. Existing studies evaluate the robustness of payload recovery and localization under image transformations, but they often focus on their own proposed method, resulting in narrow evaluations with inconsistent choices of transformations,
datasets, and metrics. These inconsistencies across studies limit direct comparisons across methods and muddle the overall picture of local watermark robustness. To address this gap, we present the first systematic robustness
benchmark for local watermarks across 55 image transformations, including (i) signal distortions, (ii) changes in image coordinate
alignment, (iii) indirect local edits, and (iv) direct watermark edits. The
benchmark evaluates MaskWM, WAM, OmniGuard, TrustMark, and PixelSeal, all methods that either provide native localization or require minimal adaptation to support it. Our results show that all evaluated methods are vulnerable to some transformation, with MaskWM standing out as offering the strongest payload recovery and localization, although it has the lowest image quality in the clean setting. Synchronization further improves MaskWM's payload recovery under several
geometric transformations, albeit at an additional cost to image quality. A key finding is that local watermark robustness depends strongly on the nature of the transformation: signal distortions are often tolerated by the strongest methods, while
geometric misalignment and generative local edits, such as inpainting and outpainting, can completely impair payload recovery. We observe that payload recovery and localization are related but not interchangeable, and both strongly depend on the transformation's impact on the watermark region.
Read the full abstract ▾
All authors Kai Yao, Bence Szilágyi, Sebestyén Kamp, Máté Poór, Máté Szilveszter, Matyas K. Zsoldos, Marc Juarez
Artificial Intelligence
Execution Flexibility in Automated Planning: A Comparative Evaluation of Deordering and Reordering Strategies
Md. Monjurul Islam, Sabah Binte Noor, Fazlul Hasan Siddiqui et al. (+1 more)
This study covers foundational concepts for enhancing plan-execution flexibility, including partial-order planning, the producer-consumer-threat formalism, and a range of deordering and reordering strategies. Creating a partial-order plan from a sequential one by removing unnecessary ordering constraints is a practical way to improve execution flexibility, and several methods have been proposed for this task. This study analyzes their capabilities across ordering, action handling, parameter handling, plan structure, concurrency, and complexity, and evaluates them against each other on a shared
benchmark. The central finding is that block deordering-based approaches, which restructure causal dependencies through block-level grouping and subplan substitution, substantially outperform MaxSAT-based approaches despite the latter's theoretical guarantees of minimum reordering. The reason is structural: minimum reordering optimizes within the causal structure already present in the plan, whereas block deordering-based methods change that structure, exposing orderings that would otherwise appear necessary. A further distinction is practical: block deordering-based methods are anytime
algorithms that always return a valid result, while MaxSAT-based methods fail entirely on a substantial portion of plans and offer no partial solution when they do. Block substitution further extends the parallel execution by formalizing non-concurrency constraints, though its impact is limited to domains with resource-based interactions. On efficiency, block deordering-based approaches achieve the highest flex gain per unit of computation time, while MaxSAT-based encodings incur large computational overhead.
Read the full abstract ▾
All authors Md. Monjurul Islam, Sabah Binte Noor, Fazlul Hasan Siddiqui, Gahangir Hossain
Artificial Intelligence
Can We Do Interpretable NLI with Graphs Based on Atomic Propositions?
Younes Boufouss, Luc Pommeret, Thomas Gerald et al. (+2 more)
While Large
Language Model (LLM)-based Natural
Language Inference (NLI) systems achieve high accuracy, their decision-making processes lack auditable structures. This paper explores whether NLI can be performed using only interpretable, graph-based representations of evidence. We introduce a fully graph-based pipeline where the classifier never directly processes the input text. Instead, sentences are decomposed into atomic propositions, converted into ConceptNet triples via constrained decoding, and represented as three graphs per pair: premise, hypothesis, and a retrieved ConceptNet subgraph. These graphs are then fed into a fine-tuned 0.8-billion-parameter
language model. On the SNLI
dataset, our pipeline achieves 89.7% accuracy, just 1.9 points below an identically trained text-based model. On ANLI, it matches the published performance of RoBERTa-large on rounds R2 and R3 (48.0% vs. 48.9% and 44.9% vs. 44.4%) but trails by 16 points on R1, resulting in an overall gap of 9 to 14 points compared to its text counterpart. We term this gap the price of interpretability and demonstrate that it stems from representational limitations rather than data constraints. Ablation studies further reveal that graphs and text are complementary: combining both modalities achieves 92.1% accuracy on SNLI.
Read the full abstract ▾
All authors Younes Boufouss, Luc Pommeret, Thomas Gerald, Patrick Paroubek, Sophie Rosset
Artificial Intelligence · Cross-listed from cs.LG
SOTER: A Generative Time-Series Foundation Model for Wearable Human Physiological Signals
Fangke Chen, Sirry Chen, Wei Chen et al. (+1 more)
Time-series foundation models have demonstrated strong cross-domain transfer, yet their common architectural assumptions remain poorly aligned with wearable physiological signals, which are multichannel, irregularly sampled, noisy, and governed by coupled continuous-time dynamics spanning distinct spectral scales. We present SOTER, a generative foundation model for wearable physiological time series that unifies cross-channel coupling, spectrum-guided expert specialization, and continuous-time latent
evolution within a single pre-training framework. SOTER combines a spatial feature-aware backbone that models inter-signal dependencies, a power spectral density (PSD)-guided mixture-of-experts layer that routes representations to experts associated with fixed spectral bands through an inspectable, non-learned rule, and a neural controlled differential equation decoder that supports prediction and imputation at arbitrary timestamps. We pre-train SOTER on 226 billion time points from five public physiological
datasets and evaluate the same pre-trained model across out-of-distribution zero-shot
forecasting, frozen-encoder linear-probe classification, and continuous-time imputation on wearable
benchmarks. SOTER achieves the best RMSE on 4 of 6
datasets and the best MAE on 5 of 6 in zero-shot
forecasting, the highest average Macro-AUROC in classification, and the lowest imputation error on all six
datasets at 75% missingness. It further remains robust to additive acquisition noise, matching or surpassing baselines evaluated on clean inputs even under the strongest corruption. These results indicate that domain-specialized foundation models for wearable physiology benefit from jointly modeling channel structure, spectral scale, and continuous-time dynamics.
Read the full abstract ▾
All authors Fangke Chen, Sirry Chen, Wei Chen, Zhongyu Wei
Artificial Intelligence
Layers, Sinks, and Scaling: Adaptive Evidence Selection for Multimodal Large Language Models
Zhenbin Wang, Lei Zhang, Lituan Wang et al. (+3 more)
Multimodal large
language models (MLLMs) can answer knowledge-intensive visual questions by combining visual evidence from images with facts retrieved from external sources. However, MLLMs may overlook relevant evidence in both modalities, attending weakly to the textual sentences or visual regions needed for the correct answer. Recent efforts address this by highlighting retrieved text and marking visual regions before generation, but apply a fixed, one-shot policy that cannot adapt to three sources of variation: whether highlighting is necessary, how much evidence different examples require, and when different textual evidence becomes relevant as the answer unfolds. We introduce Adaptive Relevance-guided Evidence Allocation (AREA), a training-free inference-time method that formulates evidence highlighting as adaptive allocation. AREA generates a single probe
token to read visual and textual relevance from fixed backbone layers, then makes three decisions: i) whether to intervene (controlled by natural
attention coverage and visual sink contamination), ii) how much evidence to expose (determined by relevance
entropy), and iii) when to refresh text during generation (triggered by causal context-
attention peaks). Across four KB-VQA and seven standard
multimodal benchmarks with nine frozen MLLM checkpoints, establishes the best performance among training-free highlighting methods.
Read the full abstract ▾
All authors Zhenbin Wang, Lei Zhang, Lituan Wang, Wei Huang, Yan Wang, Zhenwei Zhang
Artificial Intelligence · Cross-listed from cs.HC
Available but Unclaimed: An Empirical Study of Human-AI Synergy
Robin Welsch, Michelle Rausch, Pascal Knierim et al. (+4 more)
People increasingly reason with large
language models (LLMs), yet complementary capabilities do not guarantee outperforming both components. In a between-subjects study, participants (N=535) solved a 40-item battery of
matrix reasoning, mental rotation, syllogisms, and letter-string analogies, unaided or with GPT-5.6-Luna, Claude Opus 4.8, Gemini 3.6 Flash, or Kimi K3. Each assisted trial required consultation with the model. Each model answered every item alone 100 times under matched elicitation. The assisted-unaided accuracy difference increased with item-level LLM competence. Deference varied across tasks and increased with competence within tasks. Post-advice confidence distinguished correct from incorrect answers less strongly than unaided confidence. In a reference comparison, about half the increase in LLM accuracy carried through to assisted accuracy. How much of that accuracy gain reached participants differed across the models. These findings motivate evaluating LLMs in interaction with humans and designing support for selective deference that preserves independent
reasoning.
Read the full abstract ▾
All authors Robin Welsch, Michelle Rausch, Pascal Knierim, Thomas Kosch, Jochen Kuhn, Albrecht Schmidt, Daniela Fernandes