Computation and Language
Vishing-Tactics-Bench: Forecasting Exploitation Trajectories in Voice Phishing Calls
Jeongmin Lee, Dongmyung Sul, Seung Yun et al. (+1 more)
Voice phishing (vishing) unfolds in real time; by the time a call has ended and post-hoc classification is possible, the harm has already been done. The more actionable question is which concrete harm (Information Gathering or Financial Exploitation) an ongoing call is tactically progressing toward. We present Vishing-Tactics-Bench, a
benchmark grounded in Endsley's situation-awareness (SA) framework that recasts vishing defense from after-the-fact fraud classification to harm projection: predicting at each turn whether the call will reach either terminal harm. We adapt MITRE ATT&CK to vishing as a 6-tactic taxonomy (Vishing-Tactics) and label 35,340 scammer utterances across 5,645 synthetic Chinese calls. We define Exploitation Trajectory
Forecasting, a survival-style protocol over the two terminal harms with three metrics: AP@k, C-index, and divergence error. Baselines ranging from a Markov heuristic to fine-tuned LLMs show that the tactical trajectory serves as an interpretable representation of the call's tactical state, supporting harm-specific
forecasting, which can then be used for the downstream application of intervention selection; a stratified lead-time analysis at a tight false-alarm budget further identifies at what point in a call the trajectory signal yields early warning.
Read the full abstract ▾
All authors Jeongmin Lee, Dongmyung Sul, Seung Yun, Jinxia Huang
Information Theory
Conditional Fisher-Information Central Limit Theorems under Log-Concavity with Information-Theoretic Consequences
Tong Ye and Liu-Quan Yao
We establish conditional central limit
theorems in Fisher information under log-concavity in every fixed dimension. For conditionally centered normalized sums, after whitening by the averaged conditional covariance, the averaged conditional Fisher information converges to the dimension if and only if it is finite at one convolution level. The scalar criterion follows as the one-dimensional case; we also provide an independent scalar proof based on a second-order continuity
theorem for Fisher production on Gaussian-smoothed, tail-controlled classes. For the original sums, the averaged Fisher information
matrix converges in operator norm to the inverse averaged conditional covariance. Consequently, the conditional relative Fisher information with respect to the limiting Gaussian law vanishes, and the Gaussian logarithmic Sobolev inequality yields convergence in conditional relative
entropy and conditional
entropy. We give two operational consequences. For any fixed finite-constellation low-power input, the first-order conditional mutual-information slope converges to the Gaussian-noise
benchmark. For Gaussian signaling at any fixed signal covariance, the mutual-information gap from that
benchmark is bounded by the conditional relative Fisher deficit and hence vanishes asymptotically.
Read the full abstract ▾
Machine Learning
Stable-MM-R1: Anchoring Multimodal Reasoning Dynamics via Entropy-Guided Stratification
Yimeng Ye, Shuang Chen, Wenxuan Huang et al. (+8 more)
While Reinforcement Learning (RL) effectively incentivizes
reasoning in Large
Language Models, current pipelines are hindered by training instability and rapid
entropy collapse. These limitations often stem from "Rollout Silencing" and low-quality gradient signals in standard sampling procedures. In this work, we propose a robust, data-centric framework to stabilize RL training. We first introduce Potential-Aware Query Mining (PAQM), which filters data dynamically to focus on the "
Distillation Zone"—samples with high potential for capability elicitation. Furthermore, we present Hybrid Stratified Replay (HSR), a novel mechanism that restructures batches by stratifying rollouts based on Path
Entropy, a rollout-level confidence proxy, and outcome reward. Within each optimization step, HSR reuses current-policy "Stability Anchors" and "Hard Negatives" to construct high-contrast optimization groups, then clears its buffers before the next step. This approach mitigates
entropy collapse while improving the utilization of learning signals under limited compute. Our method outperforms strong baselines on complex
reasoning tasks, offering a principled solution for stable and efficient RL fine-tuning.
Read the full abstract ▾
All authors Yimeng Ye, Shuang Chen, Wenxuan Huang, Manyuan Zhang, Kaituo Feng, Zhangquan Chen, Jiayu Chen, Yucheng Zhou, Yicheng Xiao, Zhiyuan Feng, Tianyu Shi
Machine Learning
Fine-grained Distributed Backdoor Attacks in Federated Learning
Jian Wang, Hong Shen, Wei Ke et al. (+1 more)
Federated learning, as a
privacy-preserving distributed machine learning paradigm, faces significant threats from
backdoor attacks. Compared to centralized attacks, distributed
backdoor attacks are more harmful but require more poisoned samples to compensate for the loss of trigger strength due to decomposition. Fixed trigger patterns are also easily detected by robust aggregation
algorithms, increasing the risk of attack exposure. To address these challenges, we propose a fine-grained distributed
backdoor attack framework (FDBA). This framework uses dynamic trigger generation and
embedding vector optimization to perform attacks with fewer poisoned samples. First, we design a dynamic trigger generation method based on image edge structures using the Canny
algorithm to extract edge features, which are then injected with
Laplacian noise. RGB channel decomposition is applied for covert adaptation of the distributed trigger, reducing detection chances. Second, we introduce an
embedding vector contrastive learning strategy that forces poisoned samples to approach the target class center in the feature space, enhancing attack effectiveness. On CIFAR-10, piecewise-linear estimates for target ASRs between 70% and 90% show that FDBA reduces the required poisoning ratio by 37.4%–48.4% compared with DBA. In non-independent and identically distributed (Non-IID) scenarios, FDBA retains 84.7% of its IID attack performance under extreme heterogeneity, whereas DBA drops to 73.5%, and the framework successfully bypasses mainstream defense mechanisms. This study offers new insights into
federated learning security and emphasizes the potential threats and defense challenges posed by fine-grained distributed attacks.
Read the full abstract ▾
All authors Jian Wang, Hong Shen, Wei Ke, Xue Hua Liu
Robotics
EquiGQNet: Fast Grasp Quality Evaluation via Shared Equivariant Point Cloud Encoding
Sungwon Seo, Jaeseog Won, Jiyou Shin et al. (+5 more)
Planning six-degree-of-freedom (6-DoF) grasps for unseen objects in cluttered tabletop scenes from a single-view depth image requires accurate and efficient evaluation of diverse grasp candidates. Existing early-fusion methods capture local object geometry relative to each grasp candidate but repeatedly encode the scene, whereas late-fusion methods reuse a shared scene representation but may lose this grasp-relative local geometry. We propose EquiGQNet, an efficient 6-DoF grasp quality evaluator that combines the strengths of both approaches. For grasp orientation, EquiGQNet replaces the early-fusion operation of rotating and re-encoding the point cloud for each grasp candidate with an SO(3)-equivariant encode-once-then-rotate scheme, yielding grasp-aligned
geometric features from a shared scene encoding. For grasp translation, Mid-level Action Fusion (MAF) injects the grasp position into intermediate features before global aggregation, retaining local geometry relative to each candidate. We evaluate EquiGQNet in two grasp planning pipelines: Cross-
Entropy Method (CEM)-based continuous grasp search and candidate ranking with a pretrained generative planner. In simulation, EquiGQNet achieves grasping performance comparable to the early-fusion baseline and substantially outperforms late fusion on objects with complex geometry and limited graspable regions, while reducing CEM planning time from 3.31s to 0.48s, a 6.9x speedup over early fusion. In real-world household-object decluttering, EquiGQNet achieves a 95.2% grasp success rate and 230 picks per hour, versus 153 and 170 for early- and late-fusion baselines. Code is available at https://equigqnet.github.io/.
Read the full abstract ▾
All authors Sungwon Seo, Jaeseog Won, Jiyou Shin, Youngjin Seo, Hyunjun Kim, Seokmin Yoon, Tuan Luong, Hyungpil Moon
Information Retrieval
EAGER: Enrich-and-Align Generative Query Recommendation from Clicked Items in E-commerce Search
Shuwei Yuan, Mingqian Ding, Luxin Liu et al. (+2 more)
E-commerce platforms increasingly display clickable query suggestions alongside items in the user feed, enabling users to refine or expand their intent without manually reformulating queries. Existing approaches either mine suggestions from historical logs – limited to past behavior and blind to long-tail, personalized intents – or rely on off-the-shelf LLMs whose lack of platform-specific knowledge yields fluent but generic queries disconnected from real click behavior. We propose EAGER (Enrich-and-AliGn gEnerative Query Recommendation), a two-stage framework for generating query suggestions from clicked items. In the enrichment stage, supervised fine-tuning (SFT) follows a four-stage curriculum that scales information richness (from item-only to user-conditioned) and
reasoning depth (from direct to chain-of-thought). Each stage incorporates rationale augmentation, diversity regularization, and self-
distillation. In the
alignment stage, we post-train via GRPO with a hybrid reward of multiple rule-based business signals and a preference-aware click reward. Extensive offline experiments and online A/B test demonstrate the effectiveness of EAGER, which has been deployed in production at a major e-commerce platform.
Read the full abstract ▾
All authors Shuwei Yuan, Mingqian Ding, Luxin Liu, Rong Xiao, Xiaoyi Zeng
Social and Information Networks
How Well Do LLMs Simulate Survey Responses Following a Breast Cancer Screening Intervention?
Kenneth Koh, Ryan Jak Yang Lim, Alessandro Sparacio et al. (+5 more)
Collecting survey data is laborious and limited by
privacy constraints. Large
language models (LLMs) have shown promise as predictive social simulations. It is unclear whether they can replicate population-level response distributions before and after a healthcare intervention. Using information derived from 4125 women aged 35-59 years, we evaluate whether
agents informed solely by pre-intervention profile information can reproduce post-intervention response distributions. Groups of LLM
agents (n=50) were created with Gemma 4 E4B and Qwen3.5 9B; conditions ranged from zero-shot prompting to
agent profiles enriched with aggregate or individual-level demographic characteristics and pre-intervention questionnaire responses. We compared predicted and observed response distributions with Total Variation Distance (TVD) and Normalized Wasserstein Distance (NWD). Across both LLMs, profile-based
agents improved distributional accuracy relative to zero-shot and random baselines. Nevertheless, direct sampling of 50 real participants remained more accurate. Prediction errors were also higher among participants aged 55-59 years and those living in private property. Errors also varied by question theme and LLM model, with the highest errors observed for cancer fatalism and post intervention attitudes toward genetics. Sensitivity analyses showed that performance was influenced by
prompt template changes and temperature hyperparameter. Our results show the potential of LLM-based
agents to model behavioral responses to interventions in silico. However, profiles containing additional information beyond demographics did not consistently outperform simpler ones. Certain cultural constructs and population groups also remain inadequately represented by the LLM models evaluated. Future work may include building behaviorally grounded and locally validated virtual populations.
Read the full abstract ▾
All authors Kenneth Koh, Ryan Jak Yang Lim, Alessandro Sparacio, Peh Joo Ho, Mile Sikic, Borame L Dickens, Mikael Hartman, Jingmei Li
Artificial Intelligence
Encoded Early, Used Late: Where Transformers Begin to Act on an Inferred Partner's Expertise
Mika Okamoto and Gabriele Sarti
A
transformer can make an attribute linearly decodable in its residual stream at a depth where that attribute does not yet influence the output. This gap between where information is readable and where it is used has been shown for attributes stated directly in the input. We ask whether it also holds for an attribute the model must infer gradually over a conversation, namely how expert its dialogue partner is. Using ExpertCollab, a corpus of multi-turn research-planning dialogues between model-played personas at four expertise levels, we find that partner expertise is most decodable in the early layers and falls to near chance before the midpoint of the network. Counterfactual patching shows that injecting the expertise difference at the layer of peak decodability barely changes a fixed late-layer readout, whereas the same difference injected past the midpoint propagates almost completely, a separation of more than an order of magnitude. A content-matched random control and a probe-free diagnostic place the transition at the same early layer, and a statically specified control attribute stays decodable throughout. An inferred relational attribute is therefore represented well before it becomes causally active, which bounds where any attempt to read out or steer partner-conditioned behavior must intervene. We use one model on a synthetic corpus as an initial demonstration.
Read the full abstract ▾
Computer Vision
Flow3D-OPD: Multi-Teacher On-Policy Distillation for 3D Geometry Generation with Flow-Matching Diffusion Transformer
Zhiwei Ning, Zhen Zhou, Puhua Jiang et al. (+7 more)
Recent image-to-3D generation models built on flow-matching
diffusion Transformers (DiT) can produce high-fidelity meshes, yet their post-training strategy remains largely unexplored. There exist several critical bottlenecks in reinforcement learning: the inherent difficulty of defining comprehensive rewards for 3D
geometric quality, and the gradient interference that arises when jointly optimizing heterogeneous objectives. Inspired by the practicability of on-policy
distillation (OPD) in large
language models and image generation, we propose Flow3D-OPD, a two-stage post-training framework that introduces multi-teacher
distillation into 3D geometry generation. In the first stage, we utilize the semi-policy to enhance the foundational capability of the pretrained model and then design an agentic verifier for 3D
geometric quality evaluation. Based on the verifier, we could cultivate domain-specialized teacher models via direct preference optimization (DPO). In the second stage, we consolidate heterogeneous expertise into a unified student model through on-policy
distillation with hard task-routing sampling and gradient accumulation, which could mitigate the gradient interference in joint optimization. Without relying on elaborate modifications, our straightforward yet effective design achieves consistent improvements across all
geometric quality dimensions and surpasses all teacher models in the average metric. Extensive experiments demonstrate that our approach provides an effective paradigm for reinforcement learning in 3D generation.
Read the full abstract ▾
All authors Zhiwei Ning, Zhen Zhou, Puhua Jiang, Xintong Han, Gengming Zhang, Jie Yang, Zhonglong Zheng, Yuanjie Zheng, Wei Liu, Chunchao Guo
Game Theory
The Fine-Grained Complexity of Approximate Nash Equilibrium and Free Games
Noah Golowich
We study the fine-grained complexity of computing approximate Nash
equilibria and approximating the value of free games in the regime where the approximation error vanishes. Under the PCP for PPAD and ETH for PPAD
conjectures, we show that computing $\varepsilon$-approximate Nash
equilibria in 2-player $N$-action normal-form games requires time $N^{(\log(N)/\varepsilon^2)^{1-o(1)}}$, thus showing that the classical Lipton-Markakis-Mehta
algorithm (2003) is optimal through all regimes of $\varepsilon = \omega(1/\sqrt{N})$. While such optimality was known in the constant-$\varepsilon$ regime (Rubinstein, 2016), previous work could only rule out significantly smaller running times of $N^{O(\log(N)/\varepsilon)}$ in the regime $\varepsilon = o(1)$. Using similar techniques, we then establish an analogous tight lower bound of $N^{(\log(N)/\varepsilon^2)^{1-o(1)}}$ under ETH for $\varepsilon$-additive value estimation in free games, when $\varepsilon \geq 2^{-o(\sqrt{\log N})}$, answering a question of Aaronson, Impagliazzo, and Moshkovitz (2014).
Read the full abstract ▾
Computer Vision
NutriBench-Kitchen: Benchmarking Embodied AI for Nutrition Management
Yulin Wei, Xiangchen Wang, Jianhui Pan et al. (+5 more)
An
embodied kitchen assistant must do more than recognize food in isolated frames. It must track ingredient states over time and integrate visual observations with recipe and nutritional knowledge to support constraint-aware decision-making. We formalize this capability as
Embodied Nutrition Management: perceiving nutrition-relevant events, maintaining a persistent food state, and using it for knowledge-grounded planning. Existing
benchmarks evaluate static food understanding or
embodied cooking actions, but do not measure whether an
agent can continuously update and use nutrition-relevant states in dynamic kitchens. To fill this gap, we introduce NutriBench-Kitchen, a
benchmark containing 1,500 manually verified question–answer pairs from 160 cooking
videos. It covers five task families: Ingredient Entry,
Memory Management, Recipe Query, Long-Term Planning, and Short-Term Planning, spanning food-state construction, maintenance, knowledge
retrieval, and decision-making across different planning horizons. Evaluations of proprietary and open-source large
vision-
language models reveal a substantial gap from human performance, particularly in quantitative ingredient estimation, long-term state tracking, and
reasoning under interacting constraints. We further introduce Nutri-Vgent, a diagnostic long-
video agent with separate episodic, food-state, and recipe
memories. Its consistent improvements demonstrate the value of explicit state representations and structured
memory for nutrition management. Together, NutriBench-Kitchen and Nutri-Vgent provide a testbed for studying persistent state tracking and knowledge-grounded
reasoning in dynamic kitchens. Code is available at https://github.com/V1ol1n/NutriBench-Kitchen.
Read the full abstract ▾
All authors Yulin Wei, Xiangchen Wang, Jianhui Pan, Jinyu Xiao, Zheng Tan, Ruozai Tian, Guanhua Chen, Feng Zheng
Computation and Language
Retrieval-Augmented Multi-Prompt Ensemble for Minor-Grain Breeding Information Extraction
Hang Zhao and Jiahao Wang
This paper presents our system for CCL2026-Eval Task 5: Minor-Grain Breeding Information Extraction (MGBIE), which jointly extracts 12 entity types and 6 relation types from minor-grain breeding literature. We propose RAME (
Retrieval-Augmented Multi-
Prompt Ensemble), a training-free framework that elicits multiple LLM outputs under controlled diversity and aggregates them by majority voting to obtain high-confidence predictions. RAME combines (i)
retrieval-augmented few-shot selection via a hybrid BM25-
embedding retriever, (ii) a three-
prompt ensemble (Strict, Relaxed, Balanced) spanning the precision to recall spectrum, and (iii) large-scale repeated sampling with majority voting to filter noisy predictions. Built on DeepSeek-V4-Flash, RAME achieves a Total Score of 0.499 (NER 0.730, RE 0.346) on the leaderboard, ranking 1st and surpassing the official Track-A baseline powered by GPT-5.5 (0.448), representing an 11.4% relative improvement. Code is available at https://github.com/king-wang123/CCL26-RAME.
Read the full abstract ▾
Cryptography and Security
AgentLeak: Cloning Stronger LLM Agent Capabilities onto Weaker Agents Beyond Skill Stealing
Xiaoting Lyu, Yuhong Wu, Yufei Han et al. (+6 more)
Large
language model (LLM)
agents increasingly achieve long-horizon tasks by combining foundation models with explicit skills and implicit procedural knowledge acquired through execution. The resulting task-solving capabilities have become valuable proprietary assets, raising a new security question: can a substantially weaker attacker-controlled
agent acquire the capabilities of a stronger proprietary
agent through limited black-box interaction? Existing skill-stealing attacks recover explicit skill artifacts, yet we show that artifact leakage does not necessarily transfer capability: a weaker
agent may possess the same skills but still fail because it lacks procedural behaviors implicitly realized by the stronger
agent. Our key insight is that the skill execution gap itself forms a leakage surface, where missing behaviors are exposed through observable differences between successful victim executions and failed attacker executions. Based on this, we present AgentLeak, a black-box capability-cloning attack that identifies capability-critical behaviors from these execution differences and incorporates them into attacker-side skills, while keeping the attacker's model, harness, and tools unchanged. Across 20 task scenarios comprising 600 instances, diverse
agent systems, and multiple backbone models, AgentLeak improves task pass rates by over 40% compared with direct skill reuse and recovers more than 80% of the victim–attacker capability gap. Our findings reveal a confidentiality risk in LLM
agents: protecting explicit artifacts alone is insufficient, as observable execution behavior can leak the procedural knowledge required to reconstruct proprietary task-solving capabilities in low-capability and attacker-controlled
agents.
Read the full abstract ▾
All authors Xiaoting Lyu, Yuhong Wu, Yufei Han, Shichang Liu, Liang Zhang, Bin Wang, Bin Wang, Xiaobo Ma, Wei Wang
Computation and Language
Line-Coupled Language Model
Shiyuan Li, Shaorong Zhang, Zhaorui Yang et al. (+3 more)
Autoregressive
language models generate one
token per decoding step, limiting the useful output of each forward pass. Although
diffusion models, insertion-based decoding, and multi-
token prediction enable parallel generation, they either incur additional training-time
token traffic or struggle to predict strongly dependent future
tokens. We introduce the Line-Coupled
Language Model (LCLM), an autoregressive model that advances multiple text lines together by predicting the next
token for every active line while coupling the lines through shared causal context. LCLM interleaves line
tokens into a single causal sequence and uses line-staggered rotary positions, retaining the standard next-
token objective and causal
attention. Controlled experiments show that cross-line targets are substantially less dependent than consecutive same-line targets, supporting lines as parallel generation units. With 881M parameters, LCLM produces an average of 2.94 content
tokens per forward pass with a validation cross-
entropy loss of 2.44, compared with 1.00
token per forward pass and a loss of 2.39 for the vanilla autoregressive baseline. Most notably, even when LCLM generates 16
tokens per forward pass, its loss is only 0.09 higher than that of the vanilla autoregressive baseline (2.34 vs. 2.25).
Read the full abstract ▾
All authors Shiyuan Li, Shaorong Zhang, Zhaorui Yang, Qian Zhang, Greg Ver Steeg, Bingyao Li
Artificial Intelligence
EEG-Driven Decoding Framework for Passenger Hazard Perception in Highly Automated Vehicles
Yingkai Yang, Ashton Yu Xuan Tan, Bowen Li et al. (+10 more)
Reliable risk assessment remains a central challenge for Autonomous Vehicles (AVs). Despite advances in automation, passenger cognition provides a non-intrusive auxiliary signal that improves both objective and perceived safety without requiring active human intervention. We introduce an Electroencephalogram (
EEG)-based
Brain-Computer Interface (BCI) that decodes passenger neural responses for both Risk Prediction (RP) and Danger Identification (DI), explicitly modeling humans as passengers to match real-world AV use. To achieve this, we propose the Passenger Cognitive Model (PCM), Risk-aware Sequential Labeling (RSL), and the Passenger
EEG Decoding Strategy (PEDS), which integrates a 3D Convolutional Recurrent Neural Network (3D-CRNN) model for joint
EEG decoding. Experimental results show that 3D-CRNN achieves a Balanced Accuracy (BA) of $95.3\% \pm 2.7\%$ in RP and improves single-subject DI from $80.9\% \pm 3.9\%$ to $85.0\% \pm 3.2\%$ with RSL. Event-wise analyses further show that 3D-CRNN consistently outperforms other models across different event types in RP and DI. In generalization experiments, 3D-CRNN achieves $77.0\% \pm 5.3\%$ BA in cross-session DI and $77.4\% \pm 1.1\%$ BA on seen subjects in cross-subject evaluation, while maintaining a $64.9\% \pm 8.5\%$ BA on unseen subjects, demonstrating promising generalizability and transferability across both intra-subject and inter-subject variability. These findings establish an Electroencephalogram (
EEG) decoding framework for AV passenger hazard perception and suggest that passenger cognitive signals can provide auxiliary supervision for future AV decision-making and Safety of the Intended Functionality (SOTIF) support.
Read the full abstract ▾
All authors Yingkai Yang, Ashton Yu Xuan Tan, Bowen Li, Xiaorong Gao, Sifa Zheng, Jianqiang Wang, Xinyu Gu, Yang Zhao, Yuxin Zhang, Sharon X. Huang, Tania Stathaki, Jun Li, Hong Wang
Robotics
Beyond Task Success: Stage-Wise Reliability of World Model Planning under Sensing Degradation
Geonmyeong Lee and Byoung-Tak Zhang
In world model planning, sensing inputs pass through an encoder and predictor before affecting planner decisions, so final task success alone cannot reveal where sensing disturbances attenuate or persist in the pipeline. We apply 10 visual and temporal sensing degradations to a world model planner and track their effects across representation, future prediction, planner preference, and physical outcome using paired evaluation on the same 50 tasks. The relative impact of degradations was not preserved across stages: large representation shifts could attenuate downstream, while smaller initial shifts could persist to the outcome, and internal-response ordering did not directly match physical-outcome ordering. Temporal degradations also showed distinct patterns: even with similar overall changes in observation history, responses differed substantially with the location of corrupted information and the planner's actual exposure. This non-uniform stage-wise response was also observed in secondary evaluations with another manipulation task and a different world model. Stage-wise diagnosis can therefore identify where sensing disturbances attenuate or persist and help prioritize subsequent model verification and sensing mitigation.
Read the full abstract ▾
Computer Vision
A Two-Stage Framework for Ego-Centric Key Object Identification via Object State Prediction
Shihong Ling, Yue Wan, Xiaowei Jia et al. (+1 more)
This paper presents a novel framework designed to enhance key object identification in autonomous driving. Existing methods primarily focus on either detecting objects independently or leveraging visual relationships, but they do not explicitly consider the ego vehicle's perspective in determining object importance. To address this gap, we propose a structured approach that integrates a virtual ego-vehicle representation and a modular object state predictor, enabling a more accurate estimation of object behaviors relative to the ego-vehicle. Subsequently, our framework employs spatial-temporal
reasoning to refine key object identification, prioritizing objects based on their states and relative spatial information rather than relying solely on visual relationships. Experimental results on real-world driving
datasets demonstrate the effectiveness of our approach in accurately detecting critical objects in complex traffic environments.
Read the full abstract ▾
All authors Shihong Ling, Yue Wan, Xiaowei Jia, Na Du
Computational Engineering
Centering Drives Normalization Gains: Price-Offset Nuisances in Cross-Sectional Return Prediction
Mingju Chen, Qianhui Liu, Yui Lo et al. (+1 more)
Cross-sectional return prediction from raw intraday bars is sensitive to each instrument price level, an additive nuisance under a return-ranking hypothesis. We test whether removing this offset, rather than rescaling amplitudes or changing the encoder, explains gains on a point-in-time CSI~300 five-minute panel. We evaluate eight parameter-matched encoders with and without RevIN normalization; a parameter-free ladder then separates identity, scale-only, centering, last-value referencing, differencing, and standardization across all fields and restricted channels. Centering drives the reliable effect, while scale-only normalization does not help. All eight paired effects are positive and survive Holm correction on raw rank IC, after style residualization, and after further residualizing on short-term reversal. Among six stronger encoders, gains of 0.0376-0.0567 exceed the 0.0109 spread of normalized IC (0.0830-0.0939). Price-only standardization retains 93–101% of the all-field gain. These results place the main effect in transformed price-channel offset removal rather than amplitude scaling or encoder choice.
Read the full abstract ▾
All authors Mingju Chen, Qianhui Liu, Yui Lo, Yuanhang Liu
Computation and Language
PTCG: Persona-guided Tree-based Counterargument Generation
Eunbeen Son, Yohan Jo, Joonsuk Park et al. (+1 more)
The ability to generate counterarguments is important for critical thinking and balanced discourse, yet existing approaches typically produce only a single counterargument, failing to capture the diversity and persuasiveness required in real-world debates. To address this limitation, we propose Persona-guided Tree-based Counterargument Generation (PTCG), a framework that combines Tree-of-Thoughts-inspired step-wise generation and
pruning with speaker persona selection. By estimating the author's persona from the original argument and incorporating speaker personas representing distinct perspectives, PTCG operationalizes perspective-taking and enables the generation of diverse counterarguments. Results from LLM-as-a-Judge, classifier-based assessment, and human evaluations indicate that PTCG shows consistent improvements in both the diversity and persuasiveness of counterarguments compared to baseline methods.
Read the full abstract ▾
All authors Eunbeen Son, Yohan Jo, Joonsuk Park, JinYeong Bak
Computation and Language
The Illusion of Debiasing: Persona Steering Redistributes Rather Than Reduces Bias in LLMs
Ziyue Feng, Hongbo Fang and James A. Evans
Prompt-based interventions: system
prompts, personas, role instructions, reliably reshape what a
language model says, but it is unclear which layer they reach. Do they reconfigure internal structure, or only modulate the output channel? We use persona conditioning as a controlled probe, measuring its effects along a depth axis from self-report, through open-ended generation, to word-level parametric association, across three instruction-tuned models. We find a graded dissociation. Personas are legible but not structural: models follow single-trait instructions yet fail to reproduce human inter-trait covariance. The dissociation deepens with depth: personas hold or amplify closed-form QA bias, shift absolute tone while leaving between-group disparity unchanged, and barely perturb an already saturated associative baseline.
Prompt-based steering thus operates in the output channel and has a structural reach limit that surface manipulability can mask.
Read the full abstract ▾
Robotics
MAC-I$^2$: Learned Metrics-Aware Covariance for Robust Visual-Inertial Fusion in Initialization and Calibration
Xiang Fei, Yuheng Qiu, Can Xu et al. (+5 more)
Visual-Inertial (VI) fusion is fundamental to accurate and robust state estimation, where camera and IMU measurements are combined according to their respective uncertainties. Existing methods, however, fuse the two modalities with predefined uncertainties, regardless of how reliable each is in the local context, and thus often struggle under challenging environments involving illumination changes, dynamic objects, and textureless regions. In this paper, we present MAC-I$^2$, which achieves robust VI fusion through learned metric-aware covariance for both modalities, so that
vision and IMU compete on their own merits rather than relying on predefined uncertainties. Here, metrics-aware means that each predicted covariance faithfully reflects the actual magnitude of the corresponding measurement noise. On the visual side, we propagate learned feature-matching uncertainties into pose covariances for the fusion. On the inertial side, motivated by the observation that integration error accumulates sharply at the early stage and grows slowly afterward, we design a learned IMU model with a learnable initial covariance, and propose a dedicated fine-tuning strategy on a held-out training subset to enable the metrics-aware covariance on unseen sequences. As a showcase, we build a VI initialization and calibration system, since accurate and robust initialization and calibration are the prerequisite for any reliable VI system. Experiments on EuRoC, and VBR show that MAC-I$^2$ substantially outperforms existing methods: it achieves a 99.9% initialization success rate on EuRoC, reducing
gravity and velocity errors by about 60% and 42% over the strongest baseline, and maintains 80% success rate on challenging VBR sequences where baseline methods such as VINS-Mono drop below 10%.
Read the full abstract ▾
All authors Xiang Fei, Yuheng Qiu, Can Xu, Yutian Chen, Ruogu Li, Xingxing Zuo, Wenshan Wang, Sebastian Scherer
Networking
Ollama in the Wild: A Longitudinal Measurement of Exposed Ollama LLM Endpoints at Internet Scale
Zuyao Xu, Xiang Li, Yuqi Qiu et al. (+1 more)
Self-hosted large
language model (LLM) serving is emerging as a distinct category of Internet service, but we still know little about how these deployments appear and change on the public Internet. We present a 365-day longitudinal measurement of exposed Ollama endpoints (port 11434) from February 2025 to February 2026, combining daily active probing with GeoIP/ASN enrichment, PTR and port-443 host observations, and survival analysis. Across 362 observation days and approximately 4.8 million IP$\times$day observations, 26.4% of the 152,137 cumulative IPs appear for a single day; across five selected CVEs, only 0.43-2.90% of below-fix IPs upgraded in place; the top five countries/regions account for over 70% of weighted observations; and cloud and hosting providers dominate the top ASNs. These results characterize exposed Ollama as a structural exposure surface: persistent, growing, and heavily concentrated. At the same time, old versions, common model choices, cloud and hosting ASNs, PTR categories, and TLS certificate patterns remain visible across the year, indicating recurring insecure deployment practices in cloud infrastructure and the potential reach of provider-level mitigation.
Read the full abstract ▾
All authors Zuyao Xu, Xiang Li, Yuqi Qiu, Lu Sun
Systems and Control
Observer-Based Robust Control for an Aerial Manipulator System under Unknown External Disturbances
Mayank Pandey and Sneha Gajbhiye
This paper addresses the mathematical modeling and control of an aerial manipulator system comprising a quadrotor as the uncrewed aerial vehicle and a robotic arm as the manipulator. The dynamic model is established by identifying the overall center-of-mass velocity and the system's orientation as constraints, yielding a simplified, two-decoupled subsystems: a locked (overall translation) subsystem and a shape-space (actuation or overall rotation) subsystem, both subjected to external disturbances. Since the system is mechanically coupled, a critical challenge arises where constant bounded disturbances in the first subsystem manifest as time-varying, state-dependent disturbances in the second subsystem. Given that the quadrotor is inherently unstable, the movement of the robotic manipulator (RM) during flight can further jeopardize the stability of the entire QRM system if these disturbances and coupling effects are not effectively managed. To address this, we present a continuous nonlinear disturbance observer-based feedback control law, which enables the independent control of each subsystem while systematically eliminating cross-coupling effects. The efficacy of the proposed controller is validated through multiple simulations emulating practical operating conditions, thereby substantiating its real-world applicability and highlighting the core contributions of this work.
Read the full abstract ▾
Numerical Analysis · Cross-listed from math.RA
A \(3\times 3\) counterexample to Lin and Wimmer's rank-minimization conjecture associated with Roth's similarity theorem
Shuo Shi, Juan Zhang and Yun Zhang
We give a \(3\times 3\) counterexample, valid over every field, to a rank-minimization
conjecture of Lin and Wimmer (Bull. Aust. Math. Soc., 84 (3) (2011), 441–443) related to Roth's similarity
theorem for the Sylvester
matrix equation. We also prove that, over the complex field, no counterexample can occur when one of the two
matrix sizes is less than \(3\). Hence the example is dimensionally minimal over the complex field.
Read the full abstract ▾
Robotics
From LLM-Generated Specifications to Learned Quadruped Locomotion
Merve Atasever, Keyan Azbijari, Cagan Bakirci et al. (+5 more)
Quadruped
robot locomotion policies are often trained using reinforcement learning, which in turn relies heavily on hand-crafted reward functions. Designing reward functions requires substantial manual engineering, and it is often unclear which local rewards will induce the desired global behavior. Shaped rewards from formal specifications in
languages like Signal Temporal Logic (STL) can make rewards more interpretable, but writing STL specifications itself still requires domain expertise. We study whether large
language models (LLMs) can fill this gap by generating Parametric Signal Temporal Logic (PSTL) specifications that are subsequently used for policy learning. Given a natural
language locomotion objective and a constrained specification grammar, GPT-5.5 and Qwen 3.6 independently propose STL templates for command tracking, safety, and gait structure. We instantiate the parameters of the generated PSTL templates using expert trajectories and retain only specifications that are consistent with demonstrated expert behavior. The resulting specifications are then transformed into smooth, finite-history reward functions and used to train a quadruped locomotion policy with Proximal Policy Optimization (PPO) in MuJoCo XLA (MJX). We evaluate both gait-aware and gait-agnostic settings. The former specifies walking-trot, trot, and bound regimes, while the latter allows contact patterns to emerge from the task objective. We compare against hand-engineered rewards, Text2Reward-style LLM-generated reward code, and an expert-switching oracle. Gait-aware Qwen 3.6 specifications achieved 100% survival and command success across all tested speeds (0.3–2.1 m/s) and matched the target gait at high speeds, whereas Text2Reward achieved 0% for both metrics at $\geq 1.9$ m/s.
Videos: https://stl-locomotion.github.io/
Read the full abstract ▾
All authors Merve Atasever, Keyan Azbijari, Cagan Bakirci, Alfredo Reina Corona, Tolga Izdas, Richard Yang, Erdem Biyik, Jyotirmoy V. Deshmukh
Computer Vision
Discovering Natural Transformation Vulnerabilities in Black-Box Vision Models
Dongsu Song, DaeYun GO and Jay Hoon Jung
Natural
adversarial examples (NAEs) reveal that
vision models can fail under realistic semantic changes beyond norm-bounded perturbations. However, generating NAEs in a black-box setting remains challenging because existing generative attacks often rely on surrogate models, learned attack priors, or costly query-based optimization, whereas the natural transformations that expose model vulnerabilities are unknown a priori. We propose
Adversarial Scenario Attack (ASA), a query-based black-box framework that searches over natural-
language editing scenarios using a
multimodal language model and a modern text-guided generative editor. ASA jointly explores background, weather, and material/color transformations through winner–loser feedback, and uses a greedy explorer to compose only attack-improving scenarios. Across diverse ImageNet classifiers, ASA achieves substantially higher attack success rates than prior query-based generative attacks while requiring fewer victim-model queries and preserving competitive perceptual quality. Moreover, ASA exhibits both image-level and
prompt-level transferability: its
adversarial images remain effective across victim-model architectures, while its discovered editing scenarios can be reused across same-class images and, in some cases, across architectures. These findings suggest that
vision models possess reusable vulnerabilities to natural transformation patterns, which ASA can efficiently identify in a black-box setting.
Read the full abstract ▾
Robotics
Eventually Optimal and Scalable Multi-Agent Planning for Block Cave Mining
Christopher Leet, Paolo Forte, Uwe Köckemann et al. (+2 more)
Automation in underground mining has the potential to significantly enhance safety, operational efficiency, and sustainability. However, effectively coordinating fleets of autonomous vehicles in dynamic mine environments introduces substantial challenges in both optimization and motion planning. To address these challenges, we introduce and formalize the Block Cave Mining (BCM) problem, which focuses on computing a transport plan that maximizes ore throughput while satisfying draw ratio constraints. To solve this problem, we propose SAMM, an eventually optimal anytime solver that jointly integrates task assignment, scheduling, and path planning via a mixed-integer linear programming formulation. To improve scalability, we also introduce SAMMS, a variant of SAMM that trades optimality guarantees for efficiency by decomposing the problem into shorter planning subcycles. Experimental evaluations using realistic industrial mine scenarios demonstrate that SAMMS achieves near-optimal throughput and scales effectively to larger fleets and mine layouts.
Read the full abstract ▾
All authors Christopher Leet, Paolo Forte, Uwe Köckemann, Henrik Andreasson, Sven Koenig
Machine Learning
Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training
Zili Wang, Zhaopeng Qiu, Yuekai Zhang et al. (+4 more)
Speculative decoding accelerates rollout generation, which dominates the cost of reinforcement learning (RL) post-training. Online co-training keeps the draft aligned with the evolving policy, further increasing acceptance length and speedup. However, training large models on long contexts requires context parallelism (CP) and pipeline parallelism (PP), and co-training drafts under this layout poses two obstacles: (1) the draft's branch
attention is unsupported by causal CP, and (2) the target features that the draft consumes are spread across PP stages. In this paper, we present the design and implementation of online draft co-training in NeMo-RL. For CP, we extend packed, load-balanced zigzag ring
attention by merging rank-local branch
attention with causal main-sequence
attention. For PP, TapChannel transports intermediate target features across stages via a separate path, leaving the pipeline schedule unaffected. Experiments show that co-training keeps the policy's reward and accuracy close to the no-draft baseline while delivering $1.48$–$2.25\times$ rollout speedups and $1.19$–$1.88\times$ end-to-end training speedups across model scales up to 122B. Our CP design achieves strong scaling at 256K sequence length with significant
memory savings over prior work, and our PP transport incurs modest overhead. Code can be found at https://github.com/NVIDIA-NeMo/RL/issues/3698.
Read the full abstract ▾
All authors Zili Wang, Zhaopeng Qiu, Yuekai Zhang, Shuang Yu, Seonjin Na, Hayate Iso, Junjie Lai
Artificial Intelligence
Beyond Sparse Rewards: A New Benchmark and Structure-Aware Graph Alignment for Micro-Drama Understanding
Yixin Qin, Shi-Zhe Chen, Zhiqi Yu et al. (+4 more)
Micro-dramas, characterized by ultra-short durations and hyper-dense storylines, pose unique challenges for
video understanding that conventional
benchmarks fail to address. To bridge this gap, we introduce M-Drama, the first large-scale bilingual
benchmark for micro-drama comprehension, featuring over 35K instances across 9,138 clips. Furthermore, while reinforcement learning can enhance VLMs on complex narratives, existing reward metrics often suffer from sparse and superficial signals, failing to capture intricate character identities and temporal structures. We propose SAGA (Structure-Aware Graph
Alignment), a novel graph-matching reward function that models narratives as heterogeneous graphs. SAGA computes dense, rigorous rewards via decoupled semantic triplet and structural temporal matching. Extensive experiments on Qwen3-VL-8B-Instruct demonstrate that SAGA outperforms existing baselines, delivering substantial improvements in open-ended accuracy and summary quality, while maintaining competitive out-of-domain generalization. Code is available at https://github.com/qyx1121/MDrama_SAGA.
Read the full abstract ▾
All authors Yixin Qin, Shi-Zhe Chen, Zhiqi Yu, Siyuan Cheng, Tao Cheng, Jinwen Luo, Zheng Wei
Computation and Language
Revisiting Complete Reasoning Traces for Post-Training
Jaehui Hwang, Sangdoo Yun, Byeongho Heo et al. (+1 more)
Large
language models (LLMs) are often post-trained on pre-collected
reasoning trajectories to improve their
reasoning capability. Such trajectories tend to be long due to complex, interwoven paths, which often include detours on the path toward the answer. However, it has been underexplored whether LLMs indeed benefit from learning complete trajectories in post-training, such as supervised fine-tuning (SFT). Starting from our pilot study, we find that full trajectories provide only limited benefit, while partial trajectories are effective even under heavy truncation. We analyze redundancy in
reasoning trajectories through
attention-based analyses and controlled
token-removal studies, both of which show that intermediate
tokens contribute minimally to final
reasoning quality. This suggests that avoiding redundant information may allow LLMs to internally infer coherent alternatives by inferring missing steps from their internal knowledge, given known trajectory endpoints. Furthermore, we show that training LLMs using endpoints leads to consistent changes in
reasoning behavior, and that it also benefits post-training methods based on reinforcement learning or on-policy
distillation, highlighting the need to revisit complete
reasoning traces. Code is available at https://github.com/naver-ai/revisiting-trace.
Read the full abstract ▾
All authors Jaehui Hwang, Sangdoo Yun, Byeongho Heo, Dongyoon Han