Artificial Intelligence
ECMR: Evidence-Aware Deliberation Control for Multi-Agent LLM Reasoning
Yuan Liu, Haoyue Peng, Fanliang Hu et al. (+2 more)
Multi-
agent Large
Language Model (LLM) systems typically commit to a deliberation protocol before inference. A fixed protocol cannot allocate computation differently to resolved and unresolved instances, and it may mistake repeated support for independent confirmation. We study this failure as semantic false consensus: a confident group answer supported by narrow, duplicated, or correlated rationales. We introduce ECMR, an evidence-aware controller over frozen role-conditioned LLMs. At each step, ECMR summarizes answer concentration, support diversity and independence, contradictions, role behavior, verification history, and remaining budget, then selects whether to propose, challenge, verify, arbitrate, diversify, correct a role, or stop. We formulate this decision as finite-horizon budgeted control and derive sufficient conditions under which state conditioning, risk-triggered verification, and role correction improve cost-adjusted utility. Across a declared held-out evaluation of 9,000 records from MedQA, CaseHOLD, and FinQA, ECMR-Full increases pooled accuracy from 62.1% to 84.9%, reduces false-consensus rate from 40.7% to 17.0%, and raises evidence-support rate from 58.0% to 88.7%. State shuffling loses 11.7 accuracy points with only 0.7 fewer rounds, while harder traces shift from early stopping toward arbitration. ECMR-Full uses the largest
token budget, yielding the strongest absolute reliability rather than uniformly superior
token efficiency.
Read the full abstract ▾
All authors Yuan Liu, Haoyue Peng, Fanliang Hu, Lei Huang, Wenli Cai
Computation and Language
Prefilling-dLLM v2: Learned Chunk Routing for Long-Context Diffusion Language Models
Qi Han, Jing Xiong, Ziyuan Zhang et al. (+1 more)
Diffusion language models apply bidirectional
attention over the whole context at every denoising step, so a longer context makes each step substantially more expensive. Sparsifying only at inference leaves the pretrained weights unchanged, so the model is never trained under the sparse pattern it runs with. We propose Prefilling-dLLM v2, which continues training a
diffusion language model so that it routes with its native mask
token. One mask is appended at the end of each chunk, and a learned role offset marks it as a routing slot. On every fourth layer, a query keeps a bidirectional local window and the top-K chunks scored from that mask; the other layers use a bidirectional sliding window. Starting from Dream-v0-Base-7B extended to 16K with YaRN, we compare Prefilling-dLLM v2 with bidirectional NSA, DSA, and SWA under the same layout, key budget, and dense initialization, using the same Fast-dLLM decoder. With the YaRN factor scaled to the evaluation length, Prefilling-dLLM v2 leads the other sparse methods on all three RULER tasks from 16K to 128K. Its mean accuracy at 128K is 7.4 points above NSA, the next sparse model, while DSA and SWA remain near zero. At the same length, decoding is 172x faster than with full dense
attention.
Read the full abstract ▾
All authors Qi Han, Jing Xiong, Ziyuan Zhang, Ngai Wong
Machine Learning
RT-J: Large-Scale Pretraining of Relational Transformers for Context-Efficient Predictions
Rishabh Ranjan, Vignesh Kothapalli, Harshvardhan Agarwal et al. (+5 more)
Foundation models for tables and relational databases are increasingly used for predictive tasks across various scientific and industrial domains. However, adapting existing models to unseen
datasets requires tens of thousands of examples. Here we introduce RT-J, an 86M-parameter relational foundation model (RFM) that makes accurate predictions even with hundreds of in-context examples or fewer. To achieve this, we first assemble the Join, a large collection of 639 web-crawled databases with 6,023 LLM-generated forecasting tasks. We pretrain with long contexts of up to 8,192 database
cells, using in-context examples from a novel lightweight random-walk retriever. Finally, we leverage context tuning and ensembling to scale up test-time compute. On in-context regression tasks from RelBench, RT-J requires 10x fewer examples on average to match strong feature engineering pipelines with LLM-
agent or RDBLearn featurizers and TabICLv2 or LightGBM predictors. Overall, our work pushes the frontier for context-efficient predictions on relational data and highlights the importance of real-world semantics, context engineering and large-scale pretraining for RFMs.
Read the full abstract ▾
All authors Rishabh Ranjan, Vignesh Kothapalli, Harshvardhan Agarwal, Charilaos Kanatsoulis, Roshan Upendra, Tom Palczewski, Carlos Guestrin, Jure Leskovec
Computer Vision
MicroCLIP: Connecting Text and Microscopy Images via Contrastive Language–Image Pretraining
Yanran Ruan, Zhixiang Wei, Shiyao Hong et al. (+5 more)
Understanding fluorescence microscopy images is fundamental to automating biological research. However, existing methods typically learn predefined concepts from closed-set data, limiting transfer across tasks and biological contexts. In this work, we introduce MicroCLIP, a
vision-
language model that learns generalizable cellular and subcellular representations from natural
language supervision for open-set image classification and
retrieval. To pretrain MicroCLIP, we construct MicroPool-2M, a
dataset of approximately 2.5 million fluorescence microscopy image-text pairs spanning diverse cellular structures,
cell lines, and imaging conditions. Aligning microscopy images with biological concepts in natural
language is challenging because most images lack descriptive captions and biological knowledge annotations. We therefore progressively establish image-text associations at three levels to provide MicroCLIP with richer semantic supervision: (1) category-level
alignment links images to standardized labels defined by subcellular structure-
cell-line combinations; (2) knowledge-level
alignment incorporates biological priors derived from these labels to capture relationships across species and
cell types; and (3) detail-level
alignment uses a
multimodal large
language model to recaption and verify image-text pairs at scale, emphasizing fine-grained visual characteristics. Extensive experiments demonstrate that MicroCLIP achieves leading performance in zero-shot classification and text-image
retrieval on microscopy images. It retrieves images with specific cellular characteristics from thousands of candidates and generalizes to unseen biological concepts without additional supervision. These results highlight MicroCLIP's potential as a foundation model for fluorescence microscopy.
Read the full abstract ▾
All authors Yanran Ruan, Zhixiang Wei, Shiyao Hong, Xiaoxiao Ma, Yuxuan Gu, Liming Ye, Huaian Chen, Yi Jin
Machine Learning
TStruct: Learning Shared Temporal Structures for Long-Term Time-Series Forecasting
Yangze Li, Zhengyang Zhou, Qihe Huang et al. (+3 more)
Long-term time series forecasting requires temporal dependency modeling that remains stable and generalizable under temporal distribution shifts. However, real-world time series are often noisy and non-stationary, making deep learning models prone to fitting incidental temporal interactions in the training data. A key question is whether there exist temporal relationship structures that recur across different temporal positions and variables. We analyze temporal relationship
matrices on commonly used time series
datasets and find that samples from different temporal positions and different variables exhibit shared dominant temporal relationship structures. This indicates that temporal relationship structures contain patterns that recur stably across samples and variables. To learn such shared structures, we propose Temporal Structure Network (TStruct), a simple yet effective network for directly learning shared and generalizable temporal dependencies. Specifically, TStruct directly parameterizes a set of shared temporal dependency
matrices and generates normalized combination weights using dynamic sample information and static variable priors. These weights represent a soft allocation of each sample-variable pair over shared temporal dependency structures, enabling the model to generate adaptive temporal dependency
matrices under shared structural constraints, rather than fitting incidental interactions in a fully unconstrained manner. Notably, this simple structure-aware design consistently outperforms several strong baselines across all
datasets, achieving superior accuracy while maintaining high computational efficiency.
Read the full abstract ▾
All authors Yangze Li, Zhengyang Zhou, Qihe Huang, Pengkun Wang, Binwu Wang, Yang Wang
Software Engineering
State-Aware Generation of Deployable Websites from Visual Prototypes and Requirements
Ruwei Pan, Junlei Shen, Linhao Wu et al. (+7 more)
Generating deployable websites from visual prototypes and textual requirements is increasingly feasible with
multimodal models, yet keeping generated projects consistent with their prototypes remains challenging. Prototypes specify layouts and visual elements, while requirements describe interaction workflows and deployment constraints. Existing UI-to-code methods and coding
agents often treat layout reconstruction, interaction implementation, and project editing separately, allowing visual elements, files, workflows, dependencies, and render evidence to lose their correspondence across generation, verification, and update. We call this problem prototype-to-project state drift. We propose VisStateCoder, a state-aware framework that maintains these correspondences in a Project State Graph. The graph connects four types of state nodes through typed links and supports prototype grounding, graph-guided generation, and render-evidence-guided scoped updates that map failed checks to affected files while preserving files outside the update scope. We evaluate VisStateCoder on
Vision2Web using five foundation models across Static Webpage, Interactive Frontend, and Full-Stack Website tasks. Averaged across models, VisStateCoder improves overall scores over the strongest baseline for each model by 26.1 and 22.6 points on Interactive Frontend and Full-Stack Website tasks, respectively. It also reduces verified-state drift, measured by checks that pass in one update round but fail later, compared with alternative state representations. An annotation study of 50 tasks finds that the constructed graphs achieve 95.6% edge-level F1.
Read the full abstract ▾
All authors Ruwei Pan, Junlei Shen, Linhao Wu, Zixiong Yang, Qisheng Zhang, Yifan Li, Tianyu Ma, Yakun Zhang, Lu Zhang, Hongyu Zhang
Artificial Intelligence
Socratic Skill Learning: Interleaving Teaching and Practice for Transferable Agent Skills
Letian Zhu, Han Li, Yifan Yao et al. (+9 more)
Agent Skills improve LLM
agents by making task knowledge reusable across sessions. Learning transferable Skills requires evidence about a procedure and its conditions of use, which a successful task attempt can leave untested. We introduce Socratic Skill Learning (SSL), which interleaves expert teaching with Student practice. An Educator guides investigation while the Student remains the sole environment operator and Skill author, able to question, test, and adapt advice. To test reuse beyond the learning task, we construct SkillTransferBench, with 35 task families and 408 executable cases, pairing public variation specifications with held-out evaluation in fresh solver sessions. On the main 22-family, 155-case cohort, Fixed-Window SSL reaches 0.801 utility, versus 0.723 without a Skill, 0.761 for adapted Trace2Skill (trajectory
distillation), and 0.751 for adapted SKILL-KD (contrastive
distillation). Across three model pairs on shared families, interleaved guidance outperforms expert authoring and post-hoc teaching. Process analysis links guidance to Student writing and package revision; a case study shows the Student testing and refining a proposed procedure for later reuse. As demand for new lessons falls, the cost of fixed-window review grows. Issue-Guided SSL tracks unresolved questions across Skill revisions to select and focus review, reaching 0.805 utility with 61.7% fewer Educator
tokens and 16.9% fewer total acquisition
tokens than Fixed-Window SSL on the main cohort.
Read the full abstract ▾
All authors Letian Zhu, Han Li, Yifan Yao, Xinping Lei, Xinyu Che, Jiahao Wang, Dianyi Wang, Yucheng Fu, Jiakai Wang, Zhiqi Bai, Wenbo Su, Jiaheng Liu
Computer Vision
Frequency Routing and Identity Conditioning for Thermal-to-Visible Face Translation
Haojie Li, Xinlong Sun, Xiaohaoyang Lei et al. (+4 more)
Thermal-to-visible (T2V) face translation has important applications in nighttime biometric recognition, low-light surveillance, and human interpretation of thermal observations. The task remains challenging because thermal images do not directly record visible appearance, while the cross-spectral cues useful for reconstruction vary across frequency components and semantic roles. Existing spatially conditioned models tend to process these cues through a homogeneous adaptation path and do not explicitly separate local structure from global identity. We propose FreqID-Control, a parameter-efficient framework that organizes thermal guidance according to spectral reliability and semantic role while preserving the pretrained generative prior. It combines frequency-routed low-rank adaptation, which allocates separate transformation subspaces to low-frequency structure and high-frequency detail, with shared cross-spectral identity conditioning, which supplies compact global identity
tokens independently of spatial thermal
tokens. A controlled analysis on MCXFace and SpeakingFaces provides the empirical basis for this design, revealing stable low-frequency correspondence and edge-localized detail correspondence. Experiments on both
datasets demonstrate strong image fidelity and biometric identity preservation, while controlled ablations on MCXFace confirm the complementary contributions of frequency routing and shared identity conditioning.
Read the full abstract ▾
All authors Haojie Li, Xinlong Sun, Xiaohaoyang Lei, Kai Cheng, Yiguo Qiao, Xu Yang, Cunjian Chen
Machine Learning
MMTA: Benchmarking Multimodal Temporal Analysis with Time Series, Text, and Vision
Ziyang Zhang, Shenyi Li, Yilin Wang et al. (+4 more)
Time-series analysis is increasingly studied in
multimodal settings, yet most
benchmarks pair temporal signals with text or with visualizations derived from the same signal. This leaves open whether
multimodal large
language models (MLLMs) can integrate time series with real-world visual evidence. We introduce MMTA (
multimodal temporal analysis), a
benchmark of tri-modal tasks with diverse cross-modal complementarity patterns among time series, text, and
vision. MMTA spans nine domains – robotics, autonomous driving, human action, manufacturing, AR assistance, healthcare, recommendation, finance, and
climate – under
video+TS and image+TS layouts, with a 3,524-sample test split covering classification and forecasting. The
benchmark standardizes heterogeneous real-world sources into a shared schema with domain-specific
prompts, provenance documentation, and success-weighted metrics that penalize both wrong answers and failed generations. Evaluating eight open-source and proprietary MLLMs reveals a shared bottleneck: representing time series as text creates very long
token sequences and weakens fine-grained
alignment with visual evidence, especially in long-
video settings. We then introduce TimeOmni-v, a
multimodal LLM extension with a dedicated time-series encoder, temporal positional
alignment, modality-aware
token layout, and a forecasting head. TimeOmni-v improves long-
video classification and time-series inference efficiency, showing that MMTA not only exposes representation and grounding failures but also motivates targeted model design.
Read the full abstract ▾
All authors Ziyang Zhang, Shenyi Li, Yilin Wang, Ziyun Cui, Bowen Zhou, Wen Wu, Chao Zhang
Computer Vision
DVTR-KD: Distilling Vision-Language Models via Intermediate Visual Token Reconstruction
Haolan Wei, Jiahui Wang, Junli Zha et al. (+4 more)
Multimodal large
language models (MLLMs) have achieved strong performance on diverse
vision-
language tasks, but their computational demands limit deployment under constrained budgets, particularly on edge devices. Many works focus on Knowledge
Distillation (KD) and try to transfer knowledge from powerful teacher models to students. In addition to reducing model size, visual
token compression provides a complementary route to efficiency, but introduces mismatched visual
token counts that complicate direct
token-wise knowledge transfer. To mitigate this, we propose DVTR-KD, a knowledge
distillation framework based on Dense Visual
Token Reconstruction, which reconstructs the teacher's dense visual
tokens from the student's compressed
tokens. We reconstruct visual states after the final LLM layer and at the input to an intermediate layer identified as important for visual-information readout. At the intermediate layer, reconstruction is supervised through the teacher's response predictions rather than by direct
token-wise KL or MSE matching. An auxiliary
attention-readout loss complements this response-level supervision by aligning how the student and the original teacher integrate contextual information when predicting responses. Evaluations on diverse
benchmarks demonstrate that DVTR-KD improves the performance of
token-compressed students while retaining the inference efficiency afforded by visual
token compression.
Read the full abstract ▾
All authors Haolan Wei, Jiahui Wang, Junli Zha, Xinkai Lu, Yunxiao Zou, Chao Zhou, Xiang Song
Networking
QoE-Aware Service Function Chain Orchestration in Space-Air-Ground Integrated Networks for In-Flight Internet
Feng Liu, Bing Du, Ziyang Guo et al. (+2 more)
In remote regions lacking terrestrial infrastructure, aviation users increasingly demand Internet services comparable to ground cellular networks. This paper proposes a space-air-ground integrated network (SAGIN) architecture that combines ground networks, civil aviation networks, and low Earth orbit (LEO) satellites. Based on real North Atlantic flight data, a heterogeneous aviation Internet is constructed to support diverse aircraft service requirements. A composite Quality of Experience (QoE) metric, characterized by transmission delay, jitter, and content similarity, guides efficient resource utilization. The dynamic connectivity of SAGIN is modeled using a reconfigurable time-expanded graph (RTEG). With Network Function Virtualization (NFV), data delivery is formulated as a Service Function Chain (SFC) deployment across ground stations, relay aircraft, and satellites, resulting in a long-term QoE maximization problem. A tailored Multi-
Agent Deep Deterministic Policy Gradient (MADDPG)
algorithm incorporating LCSS-based path filtering and softmax-based action selection is developed. Simulation results demonstrate that the proposed approach enables seamless connectivity for heterogeneous aircraft and consistently outperforms game-theoretic and MATD3 methods in terms of QoE.
Read the full abstract ▾
All authors Feng Liu, Bing Du, Ziyang Guo, Xiaoqi Zhang, Haijun Zhang
Machine Learning
Persistence Positional Encoding: A Minimal Transformer Recipe for Persistence Diagrams
Eunwoo Heo, Ganghyun Kim and Jaejun Yoo
Persistence diagrams (PDs) represent multiscale topological information as multisets of birth-death pairs $(b,d)$. How best to use them with
Transformers, however, remains unclear. Connecting PDs to
Transformers requires choosing how topological information is tokenized and exposed to self-
attention. Prior methods use specialized designs for tokenization, aggregation,
Transformer blocks, or
attention, but are hard to compare since reported results rarely share PD inputs, splits, or model-selection rules. We instead confine the diagram-specific design to point representation, mapping each birth-death pair through fixed random
Fourier features while leaving the
Transformer backbone unchanged, and call this recipe Persistence Positional Encoding (PPE). In matched controls on eight TU graph
benchmarks (TU8), changing only point representation shifts macro accuracy by 5.49 to 6.47 percentage points (pp), against 0.47 to 1.45 pp from changing the downstream module, identifying point representation as the dominant design variable tested. We further reevaluate vectorizations, kernels, learned representations, and
Transformer-based models under one protocol with shared PD inputs, outer folds, and validation-only selection. Among all evaluated PD methods and downstream configurations, PPE achieves the highest mean on most
benchmarks, including TU8, ORBIT5K, and MCB-C. Finally, we prove that grid averaging leaves a reconstruction error at any resolution, whereas a
Transformer with fixed random
Fourier frequencies is almost surely universal. PPE thus offers a minimal,
Transformer-native baseline for persistence diagrams, and a common protocol for future comparison.
Read the full abstract ▾
Machine Learning
GeoNIRS: Geometry-Aware Pretraining for Heterogeneous fNIRS Representation Learning
Zongsheng Li, Shengyuan Zhu, Zimo Zhang et al. (+3 more)
Functional near-infrared spectroscopy (fNIRS) provides portable and spatially localized measurements of cortical hemodynamics, but differences in channel count, probe layout, signal distribution, and task type hinder joint pretraining across
datasets. We introduce GeoNIRS, a geometry-aware framework that uses source-detector locations as a shared spatial reference across heterogeneous probe layouts. GeoNIRS compresses variable-size channel sets into a fixed-size latent representation and employs bidirectional HbO/HbR cross-reconstruction with contiguous temporal masking. GeoNIRS-base is pretrained on eight heterogeneous
datasets, while GeoNIRS-large scales to 22
datasets. Neither downstream data nor labels are used during pretraining, enabling evaluation under a strict cross-task setting. To our knowledge, GeoNIRS is the first fNIRS framework to explicitly model physical channel geometry during joint pretraining across heterogeneous channel configurations, and the first to provide complete source code and downloadable pretrained checkpoints. We evaluate GeoNIRS on eight held-out tasks spanning clinical and auditory assessment, pain and affective processing, semantic decoding, and motor function. GeoNIRS achieves the highest task-averaged balanced accuracy among the evaluated conventional machine-learning, fNIRS-specific, general time-series, and fNIRS pretraining baselines. Compared with random initialization, pretraining improves average balanced accuracy by 4.12 percentage points under linear probing and 3.30 points under full fine-tuning, with larger benefits when labels are scarce. GeoNIRS also remains robust to missing input channels. Geometry perturbations and pretraining ablations further show that both channel geometry and cross-chromophore reconstruction contribute to the learned representations.
Read the full abstract ▾
All authors Zongsheng Li, Shengyuan Zhu, Zimo Zhang, Hanyao Li, Shengming Wang, Quanying Liu
Computer Vision
What Makes a Good Visual Representation for Spatial Understanding in MLLMs?
Xiaohu Huang, Yumin Guo and Kai Han
Spatial understanding is a necessary step in moving
multimodal large
language models (MLLMs) toward
reasoning about the 3D physical world. Their visual pretraining data, however, is almost entirely 2D, so the visual
tokens processed by MLLMs often lack inherent 3D structural information. A growing number of methods inject such structure by transferring representations from external
vision or geometry models, and collectively establish that such transfer helps. Behind these methods lies a largely underexplored question: what makes a visual representation good for spatial understanding? To answer it, we first separate what is transferred from how it is transferred, and then trace the resulting differences back to properties of the representations themselves. What we transfer comes from 3D geometry models, self-supervised image and
video models,
vision-
language models, and generative models, in the form of latent features or explicit depth and point maps. We transfer it either by
distillation, using it as a supervision target, or by fusion, feeding it as an additional input. We find that a representation's 3D structure correlates strongly with how much it helps, measured by cross-view consistency for latent features and by depth and reconstruction accuracy for explicit geometry. We further show that the form of a representation dictates the mechanism it needs. Latent features gain as much from fusion as from
distillation; explicit geometry gains little from fusion, and most of its
distillation gain depends on a geometry decoding head. Guided by these findings, we distill both latent features and explicit geometry into a single MLLM. Taking only RGB
video as input, it consistently improves over its base model on both static 3D and dynamic 4D spatial
reasoning benchmarks.
Read the full abstract ▾
Robotics
Physics-Aware Object-Centric SE(3) Forecasting for Real-Time Proactive Dynamic Grasping
Yirui Zhou, Hongli Xu, Ying Zheng et al. (+2 more)
Proactive dynamic grasping requires
robots to predict the future state at which physical contact will occur, rather than reacting to the currently observed object pose. But what future must the
robot predict, and at what level of abstraction? We argue that contact-relevant rigid-body state in SE(3) is a compact task-sufficient abstraction for this problem: it directly determines where and when the
robot should establish contact, without requiring visually complete future scenes. This insight leads to a lightweight object-centric forecast-then-select framework that predicts future rigid-body trajectories from short pose histories and uses a reachability-aware interception selector to choose feasible contact states under kinematic and timing constraints. Trained with randomized physics simulation over primitive shapes, the forecaster transfers to real objects once their metric geometry and SE(3) state histories are recovered. Experiments across simulation and real-world dynamic grasping tasks show that our framework improves the average success rate from \(24.5\%\) to \(62.0\%\) in simulation, and from \(15.3\%\) to \(67.3\%\) in real-world catching, conveyor, and turntable interception. The SE(3) forecaster runs in \(15\,\mathrm{ms}\), orders of magnitude faster than dense visual future prediction, enabling real-time contact decision making. Our results suggest that effective proactive grasping does not require imagining the entire future scene. Instead, predicting the compact physical future that determines contact provides an effective and practical abstraction for real-time dynamic grasping.
Read the full abstract ▾
All authors Yirui Zhou, Hongli Xu, Ying Zheng, Yi Wang, Lap-Pui Chau
Information Retrieval
FASHION-RQ: Mining Hierarchical Semantic Structure for Automated Evaluation of Fashion Retrieval
Tripti Shukla, Shwetha Somasundaram, Akshat Gupta et al. (+3 more)
Evaluating fashion
retrieval systems remains challenging due to the limited availability of
benchmarks that capture the fine-grained, open-ended queries encountered in real-world search and recommendation. Existing
benchmarks often rely on manual curation, predefined taxonomies, or synthetic
prompts, limiting scalability and semantic coverage. We introduce FASHION-RQ: Mining Hierarchical Semantic Structure for Automated Evaluation of Fashion
Retrieval, an automated framework that derives user-type queries and relevance sets directly from catalog structure. Our key insight is that hierarchical discrete representations learned by Residual Quantized Variational Autoencoders (RQ-VAE) organize products into semantically coherent clusters at progressively finer granularities. We leverage this structure to construct FASHION-RQBENCH, a scalable
benchmark without manual query annotation or predefined taxonomies. We propose four complementary tasks for evaluating ranking, relevance classification, fine-grained sibling discrimination, and paraphrase robustness, alongside query- and attribute-level relevance metrics. Experiments across recent
retrieval models demonstrate that FASHION-RQBENCH differentiates model capabilities across query types and attributes. A human study further shows that
benchmark queries achieve comparable naturalness and higher intent clarity than real user queries, supporting their validity for real-world evaluation. The code and FASHION-RQBENCH
dataset will be publicly released upon acceptance.
Read the full abstract ▾
All authors Tripti Shukla, Shwetha Somasundaram, Akshat Gupta, Rohit Garg, Paul Duff, Ian Anderson
Artificial Intelligence
Learning Adaptive Deliberation Control for Reliable Recognition with Multi-Agent Language Models
Yuan Liu, Haoyue Peng, Fanliang Hu et al. (+2 more)
Multi-
agent deliberation can improve recognition by criticizing and revising candidate answers, but fixed controls cannot respond to changes in disagreement or
reasoning quality. We present an adaptive multi-
agent controller (ADMA) that learns continuous policies for proposal, challenge, arbitration, and coordination while keeping the task model frozen. Shared trace features describe quality, agreement, and progress, while role-specific features support decentralized control during inference. One controller trained on legal and financial
reasoning is transferred without policy updates to four textual and
multimodal benchmarks and two successive task-model versions. With Qwen3.5 held fixed, ADMA-Full improves accuracy by 6.2-14.2 percentage points over Direct and by 2.8-3.0 points over ADMA-Fixed. Paired bootstrap ranges for the Full-Fixed comparison are positive on three tasks. Outcome decompositions show that recoveries exceed introduced errors, while process assessment records fewer false-consensus events and more supported correct outcomes. Transfer to Qwen3.6 remains task dependent, which bounds the generality of the learned policy. Median
token and latency ratios are 14.3-18.7 and 7.8-16.5 relative to Direct. These results show that adaptive deliberation can improve fixed-model recognition when its task-dependent gains justify the additional computation.
Read the full abstract ▾
All authors Yuan Liu, Haoyue Peng, Fanliang Hu, Lei Huang, Wenli Cai
Sound
Beyond Fixed Targets: On-Policy One-Shot Speaker Personalization for Autoregressive Text-To-Speech Models
Sangkwon Park, Che Hyun Lee, Heeseung Kim et al. (+1 more)
One-shot personalization of autoregressive (AR) text-to-
speech (TTS) models remains challenging because supervised fine-tuning (SFT) on a single reference pair rapidly degrades generation quality. We identify fixed, off-policy supervision as a key cause of failure in one-shot SFT for AR TTS, leading to blank or near-empty outputs at inference. Diversifying the training data with zero-shot pseudo-pairs eliminates this failure but yields little improvement in speaker similarity and increases word error rate because the targets remain fixed and teacher-forced. We therefore propose two on-policy approaches for one-shot speaker personalization: Group Relative Policy Optimization (GRPO) with a speaker-similarity reward and On-Policy Self-
Distillation (OPSD) with soft supervision from a reference-conditioned teacher. Both on-policy methods improve speaker similarity while preserving intelligibility, with different trade-offs: GRPO achieves the strongest overall performance, whereas OPSD enables substantially faster adaptation without external reward evaluation. Further analysis confirms that these gains stem from on-policy supervision rather than target diversity, establishing on-policy training as an effective framework for one-shot AR TTS personalization.
Read the full abstract ▾
All authors Sangkwon Park, Che Hyun Lee, Heeseung Kim, Sungroh Yoon
Cryptography and Security
ProtoArena: Can AI Agents Discover and Exploit Real-World Cross-Component Protocol Vulnerabilities?
Fasheng Miao, Zhaoyu Hu, Xiang Li et al. (+5 more)
Large
language model (LLM)
agents are increasingly used for vulnerability analysis, yet existing
benchmarks provide limited coverage of exploitation that depends on interactions among network protocol components. Such vulnerabilities arise from inconsistencies in message interpretation, timing, state, or trust assumptions across participants. We introduce ProtoArena, a black-box, multi-component, dynamic
benchmark with 80 cross-component vulnerability-exploitation environments spanning 11 protocol and application domains and 7 attack categories. Tasks are primarily informed by published protocol-attack mechanisms.
Agents interact with running systems through scoped interfaces to achieve security-effect objectives without target source code. Three information levels assess guided reproduction, exploitation with mechanism and
topology context, and effect-directed black-box exploration. A hybrid verifier combines rule-based checks with LLM review of independently collected traffic and logs. We evaluate five LLMs. Claude Opus 4.8 achieves the highest observed success rates: 48.8% under L1 guidance and 12.5% under effect-directed black-box L3. In a separate DNSBomb optimization experiment, it achieves a peak downstream-response to average query-input byte-rate ratio exceeding 171,000 times. The results show that
agents can achieve verifiable exploitation effects in some cross-component environments, but success with limited black-box information remains low and varies across domains and attack categories. All experiments use controlled, explicitly authorized targets; they do not test production systems or disclose previously unknown vulnerabilities.
Read the full abstract ▾
All authors Fasheng Miao, Zhaoyu Hu, Xiang Li, Changqing An, Jiutian Zeng, Longtao Huang, Haixin Duan, Jilong Wang
Artificial Intelligence · Cross-listed from quant-ph
A Tutorial on Quantum Amplitude Embedding with Data Re-Uploading
Hyojun Ahn and Joongheon Kim
Amplitude
embedding lets a variational
quantum model read many classical features with few
qubits, but the model output is then only a quadratic form in the normalized input. Data re-uploading, which encodes the input repeatedly between trainable blocks, can be used to enrich such a model. Because every additional upload of an amplitude
embedding repeats a costly state preparation, the gain that an upload buys has to be known before the gates are spent. This tutorial thus addresses the problem of explaining what data re-uploading adds to an amplitude-embedded model and how large the gain is. We express sequential re-uploading and parallel re-uploading with two copies in one notation and show that each upload raises the reachable
polynomial order by two. Under one trainable
circuit family and one budget of trainable parameters, simulation results show that re-uploading raises the test coefficient of determination from 0.689 to between 0.883 and 0.975 on targets of order four and from 0.337 to between 0.642 and 0.783 on targets of order six, on all ten seeds. On a quadratic target and on handwritten digits, one upload remains the best choice, and every additional upload lowers the gate error that the classifier tolerates. These results help designers match the number of uploads to the
polynomial order of the task at hand.
Read the full abstract ▾
Numerical Analysis
Learning Hamiltonians from Trajectory Data with Kernel Methods
Jianyu Hu, David Martin de Diego, Juan-Pablo Ortega et al. (+1 more)
We study the recovery of a
Hamiltonian from discrete trajectory observations using structure-preserving kernel methods. When exact-flow data are fitted through a symplectic one-step scheme, the fixed-step learning target is generally not the physical
Hamiltonian itself, but a scheme-dependent inverse modified
Hamiltonian. For the implicit midpoint and $p$-implicit symplectic Euler methods, we establish the existence of analytic inverse targets and show that their discrepancies from the physical
Hamiltonian are second and first order in the observation step, respectively. Within an RKHS framework, we derive finite-sample reconstruction bounds for random trajectories, allowing dependence among observations from the same trajectory, and separate the error into estimation, approximation, and integrator-induced components. We further establish stability under noisy trajectory observations and develop a multistride Richardson correction that systematically raises the order of the integrator-induced error without requiring additional trajectories. Numerical experiments confirm the predicted scheme-dependent orders, the effect of Richardson correction, and the applicability of the method across several
Hamiltonian systems.
Read the full abstract ▾
All authors Jianyu Hu, David Martin de Diego, Juan-Pablo Ortega, Miguel Vaquero
Data Structures and Algorithms
PHastR: Faster Minimal Perfect Hashing with Rotated Layouts
Sebastiano Vigna
We present PHastR, a minimal perfect hash function of the PHast family. Like PHast+, the fast-to-build variant of PHast, PHastR finds feasible seeds with bit-parallel operations: its seeds move keys along rings of slots, and to evaluate the seeds of a bucket it cyclically shifts, for each key, a word recording which slots of its ring are used. Unlike PHast+, however, it is almost immune to self-collisions, which make two keys of a bucket collide for (almost) all its seeds: more than a third of the buckets bumped by PHast+ with wrapping and multiplier 3 contain one. With 8-bit seeds, PHastR uses $1.94$ bits per key, essentially the same space as the best variant of PHast+ with wrapping, but it is built in $37\%$ of the time, and its queries are $28\%$ faster on $10^9$ keys, as it bumps $0.5\%$ of the keys instead of $3.9\%$; they are faster also than those of PHast, which takes twelve times as long to build, and about as fast as those of PtrHash, which uses more than one bit per key more. With a larger expected bucket size PHastR comes within $0.02$ bits per key of the space of PHast, using $1.90$ bits per key, and still has faster queries. Its construction also scales better with the number of threads: with all the threads of a 48-core machine it is almost three times faster than that of PHast+. Finally, we argue that when each key can be placed only within a range of $W$ slots, every placement of $n$ random keys on $n$ slots, however found, must bump on large key sets a constant fraction of the keys: we derive a lower bound of about $0.1/W$ on this fraction, and we estimate that the best placement bumps about $1/2W$ of the keys, which is about one tenth of the keys bumped by PHastR.
Read the full abstract ▾
Machine Learning
Agentic-MatriBO: Capability-Compiled Agentic Bayesian Optimization for Materials and Chemical Design
Yu Zeng, Yuchuan Chen, Ziyue Zhang et al. (+5 more)
Materials and chemical optimization must choose the next synthesis, characterization, or high-fidelity simulation from sequential observations under a limited budget. Large
language models can use task knowledge to propose adjustments, but composing surrogates, acquisition functions, and search strategies is constrained by compatibility, parameter dependencies, and evidence requirements. We present Agentic-MatriBO, a capability-compiled agentic
Bayesian optimization (BO) framework that represents BO components and selection conditions as composable capabilities. A
language policy proposes BO recipes and candidate plans; a compiler checks proposals against the registered capability catalog; and a deterministic execution layer validates and commits candidates. Oracle observations are stored in research
memory and inform later rounds. We compare Agentic-MatriBO with eight baselines on four offline finite candidate pools—COF, PCE10, Buchwald–Hartwig (BH), and Suzuki–Miyaura—using five seeds per condition; results are reported descriptively. Across the four tasks, Agentic-MatriBO has the best mean endpoint quality after orienting each objective so that higher is better. On COF, it reaches the tabulated optimum in all five seeds both in the end-to-end comparison and after shared initialization, where it also retains an advantage in cumulative quality. Its normalized area under the best-so-far quality curve during the BO stage (nAUC-BO) is substantially higher on COF, near parity with the best baseline on PCE10 and BH, and lower on Suzuki. COF ablations show that removing the capability compiler or evidence gate coincides with lower cumulative quality and fewer exact-optimum hits. The framework provides an inspectable interface for evidence-conditioned BO adaptation.
Read the full abstract ▾
All authors Yu Zeng, Yuchuan Chen, Ziyue Zhang, Zijian Wei, Yanyi Su, Guojiang Zhao, Hongshuai Wang, Zhifeng Gao
Software Engineering
Diff Diffusion: Predicting Edit Paths with Diffusion Models for Efficient Code Editing
Xiaolong Xu, Hao Jiang, Bo Jiang et al. (+5 more)
Diffusion large
language models (dLLMs) are appealing for code generation because they can update many
tokens in parallel, yet many are optimized for sequence synthesis rather than edit prediction. This setting is poorly matched to modern AI-assisted IDEs, where edit prediction must infer the next code change from recent edits, the current file, and cursor-local context. Such changes are usually sparse and preserve most source
tokens, so regenerating a full code segment wastes computation. We propose Diff
Diffusion, a
diffusion-inspired LLM framework for code editing that predicts edit paths from an original snippet to its revised form. Diff
Diffusion represents denoising as structured edit operations, and combines this formulation with multi-stage training and efficient parallel decoding. By focusing computation on changed regions, Diff
Diffusion supports accurate low-latency edits and multi-position parallel modification. Across public edit-completion
benchmarks, edit-prediction evaluations, and online validation, Diff
Diffusion matches or leads autoregressive baselines while delivering 2.78 times faster inference on public edit-completion
benchmarks and a 39.0% P50 model-latency reduction in production. This work provides a new modeling perspective for AI applications that require high accuracy and low latency.
Read the full abstract ▾
All authors Xiaolong Xu, Hao Jiang, Bo Jiang, Zhiwen Deng, Ningyuan Sun, Bingzhou Chen, Jin Chen, Chenxing Wei
Computer Vision
QK-Edit: Revisiting Attention-based Injection in MM-DiT for Image and Video Editing
Tiancheng Shen, Zilong Huang, Xiangtai Li et al. (+6 more)
Multimodal Diffusion Transformers (MM-DiTs) have recently emerged as a powerful framework for unified text-
vision synthesis, surpassing traditional U-Net architectures in generative tasks. One key innovation lies in its
Multimodal Self-
Attention (MM-SA) interaction, where image and text
tokens are concatenated and processed via self-
attention. However, this mechanism poses significant challenges for editing, rendering conventional U-Net-based
attention manipulation methods ineffective. To address this limitation, we propose QK-Edit, a training-free framework that exploits the unique
attention dynamics of MM-DiTs for precise text-guided image and
video editing. By introducing a novel query-key manipulation strategy, our method isolates and adjusts critical
attention components to achieve an optimal balance between
prompt fidelity and structural consistency. This enables seamless edits across various tasks, including object addition, object removal, object replacement, changing background, changing material, changing color, and style transformation. Notably, it can be easily implemented with feature replacement in inference. QK-Edit demonstrates superior editing performance on state-of-the-art models, such as FLUX and HunyuanVideo, effectively bridging the gap between generative power and editable flexibility in MM-DiTs, and paving the way for scalable
multimodal content creation.
Read the full abstract ▾
All authors Tiancheng Shen, Zilong Huang, Xiangtai Li, Zhijie Lin, Jiyang Liu, Yitong Wang, Jiashi Feng, Ming-Hsuan Yang, Jun Hao Liew
Artificial Intelligence
Benchmarking and Training Reward Models for Cross-Source Evidence Grounding in Search Agents
Haochen Zhao, Minsheng Hao, Zheng Li et al. (+8 more)
Existing approaches to training and evaluating search
agents focus heavily on final-answer quality, largely neglecting how generated answers map back to retrieved evidence. Recent methods track citations to enforce grounding, but citation support alone does not establish source independence or whether an answer appropriately handles conflicting evidence. We argue that credibility in open-web
retrieval depends on principled cross-source evidence grounding that accounts for source independence and systematically attributes and adjudicates conflicting evidence, while hierarchical judgments of these grounding decisions provide a unified signal for training and evaluation. To this end, we propose an evaluation framework, Cross-Source Evidence Grounding (CSEG), that employs a fine-grained, six-layer decision tree over the complete search trajectory. We introduce CSEG-Bench and trainable reward models to provide targeted supervision following this framework. Our approach helps eliminating the severe data skew of natural trajectories by providing targeted coverage over complex, deep-layer decision paths, such as cross-source conflict resolution and nuanced attribution, that standard rollouts fail to capture. Specifically, (1) we define the hierarchy at label, node, and path levels; (2) we construct 1,160 verified trajectories over nine canonical paths by rewriting answers on real retrieved document pools, with independent frontier judges reproducing the entire gold path and human experts deciding inclusion; (3) we train CSEG-RM, which reaches 83.9% exact accuracy on our
benchmark while lying on the performance-
token efficiency Pareto frontier. Experiments that use this reward alone to supervise a frozen-
retrieval answer policy raise the independently judged mean label from 2.83 to 4.19, and show that learning the hierarchy generalizes across layers at a fraction of frontier-judge cost.
Read the full abstract ▾
All authors Haochen Zhao, Minsheng Hao, Zheng Li, Yifan Li, Zikang Liu, Bowen Qin, Yuhao Cheng, Chuang Fan, Haosheng Zou, Jeff Chen, Sujian Li
Machine Learning
Checking the Checkers: A Synthesis of Verifier Noise and Robustness in RLVR
Savyasachi Gupta and Aniruddha Srinivas Joshi
Reinforcement learning with verifiable rewards (RLVR) has become a principal recipe for improving the
reasoning of large
language models (LLMs). It rests on the assumption that the checker scoring each response is itself correct. In practice, rule-based checkers reject correct answers in unexpected formats, unit tests accept incorrect programs, and LLM judges can be led to approve empty or
adversarial responses. This work surveys 58 studies that measure, analyze, or mitigate such verifier errors. We ask what errors verifiers make, how they affect training, which mitigations work against which errors, and how the findings extend to self-generated, rubric- and judge-based, and verifier-free rewards. The evidence suggests that the structure of an error matters more than its rate. Random errors mostly slow learning, though under a fixed budget or KL regularization they also lower final accuracy. False positives that the policy can learn to trigger can cause plateaus or collapse, sometimes even with a low initial error rate. Whether a verifier wrongly rejects or accepts depends on what it checks and who built it, not on its family. Its errors shift as the policy improves, and most mitigations have been tested against only one kind of error. We compile the findings into comparison tables, reconcile apparent conflicts in the literature, and propose a reporting protocol to make future results comparable.
Read the full abstract ▾
Information Retrieval
Mitigating Popularity Bias in Sequential Recommendations with Sparse Prior-Regularized Objective
Anton Pembek, Anton Klenitskiy, Alexander Savchenko et al. (+1 more)
Popularity bias is a well-known challenge in recommender systems, often leading to the over-exposure of popular items and reduced catalog coverage. In this work, we propose PR-entmax, a sparse prior-regularized objective that makes popularity correction context-dependent. We derive our method from a generalized variational principle in which an item popularity prior is incorporated directly into a sparse probability mapping, yielding a popularity-aware active support: the set of items receiving non-zero probabilities and hence non-zero gradients is determined jointly by contextual relevance and item popularity. As a result, the prior affects not only item probabilities, but also which items participate in gradient updates during training. The method requires no architectural changes and can be integrated into existing training pipelines by modifying only the loss function. At inference time, the prior term is removed, and items are ranked by the learned contextual relevance alone. Experiments across multiple
benchmark datasets and sequential recommendation architectures show that PR-entmax consistently reduces popularity bias while preserving or improving recommendation relevance, advancing the relevance-debiasing trade-off.
Read the full abstract ▾
All authors Anton Pembek, Anton Klenitskiy, Alexander Savchenko, Alexey Vasilev
Sound
FLOMSS: Generative Music Source Separation by Flow Matching
Shuyang Cui, Zhi Zhong, Keisuke Toyama et al. (+10 more)
Music source separation (MSS) has been conventionally addressed by discriminative methods. Recently, flow matching has emerged as a strong generative paradigm for modeling complex data distributions, with successful applications across diverse domains. These advances motivate us to systematically investigate flow matching for MSS in a four-source (Vocals, Bass, Drums, Other) setting in full-band stereo. Specifically, we investigate prediction targets, flow paths, single-source versus multi-source modeling, and mixture consistency constraints. Beyond the signal-level metrics commonly used for MSS, such as signal-to-distortion ratio, we further evaluate MSS methods using
embedding-based metrics, as well as a listening test, providing a comprehensive
benchmark. On MUSDB18-HQ, our proposed model exhibits competitive performance in both objective and subjective tests, highlighting the potential of flow matching for generative MSS. Demo: https://ik4sumii.github.io/FLOMSS/
Read the full abstract ▾
All authors Shuyang Cui, Zhi Zhong, Keisuke Toyama, Koichi Saito, Qiyu Wu, Yukara Ikemiya, Junghyun Koo, Kin Wai Cheuk, Woosung Choi, Giorgio Fabbro, Wei-Hsiang Liao, Shusuke Takahashi, Yuki Mitsufuji
Cryptography and Security
Same Attack, Different Judge: A First Controlled Adversarial Comparison of Jev, LLM Judges, and Purpose-Built Safety Classifiers
Xiang Li, Aichun Huang, Mingming Zhang et al. (+1 more)
TypeSafe AI's Jev is a new kind of safety judge. It generates no text and no label: it answers questions the developer defines in advance and returns a probability for each answer. The vendor admits that
adversarial input can change these answers but gives no numbers. In this paper, we present the first controlled
adversarial evaluation of Jev. We build a test harness that sends every system the same payload under the same policy, and run it on four public-data tasks, from rule-based ones (input moderation, injection detection, routing) to a judgment-based one (phishing detection), with fifty attack variants from thirteen families. The targets are Jev, five GPT and GLM judges, and two local classifiers (Llama Guard 3,
Prompt Guard 2). We find that Jev is the most robust system on input moderation: eight forgeries of its structured input bypass none of its decisions, and a sentence asking only why someone might make a harmful request bypasses 9% of its blocks, against 56% for GPT-6 Astra. On phishing detection. However, the same forgeries bypass 19-29% of the emails it flags, and even without attack Jev detects only 62% of phishing emails, against 77-81% for the GLM judges. Jev is also the fastest judge, but it flags the most benign requests (27%). The local classifiers fail differently: a training-example statement makes Llama Guard 3 miss 62% of the phishing it had flagged. These results suggest that Jev resists attacks on tasks with a clear rule to check but is more exposed on tasks that call for an overall judgment.
Read the full abstract ▾
All authors Xiang Li, Aichun Huang, Mingming Zhang, Zheli Liu