The paper investigates how large language models (LLMs) handle bug fixing versus problem solving in competitive programming. Using a dataset of ~3,000 Codeforces submissions and their human fixes, the authors compare LLM-generated patches to human patches and assess whether LLMs prefer to modify buggy code or generate new solutions. Results show that LLMs often alter more lines than necessary and sometimes produce entirely new solutions, performing better when allowed to generate solutions from scratch rather than patching existing code.
By Alexandru Stefan Stoica, Traian Rebedea, Marian Cristian Mihaescu
STRAND is a new benchmark that tests multimodal large language models’ ability to track objects, their states, and relationships over time in videos. It evaluates intermediate reasoning by breaking queries into sub‑questions and uses Faithful Accuracy to ensure all parts of an answer are correct. The authors also propose an object‑centric framework that builds structured trajectories and shows reduced hallucinations and better temporal consistency compared to existing models.
By Thong Nguyen, Tri Cao, Khoi Le, Cong-Duy Nguyen, Quynh Vo, See-Kiong Ng, Bryan Hooi Kuen-Yew
The paper introduces a new evaluation framework for confidence estimation in large language models, focusing on three properties: robustness to prompt changes, stability across semantically equivalent answers, and sensitivity to semantically different answers. It demonstrates that existing confidence estimation methods perform well on robustness and stability but often fail to detect differences in answer meaning, revealing gaps in current evaluation practices. The framework aims to guide the selection of confidence estimators for practical applications.
By Yuxi Xia, Dennis Ulmer, Terra Blevins, Yihong Liu, Hinrich Sch\"utze, Benjamin Roth
MoVISA introduces Multi-Token Reasoning for Video Object Segmentation, using multiple segmentation tokens (e.g., SEG0, SEG1) instead of a single token to better localize multiple objects over time. This approach enhances fine-grained alignment between language prompts and spatio-temporal mask predictions, leading to improved performance and interpretability. On benchmarks such as MeViS, DAVIS17, ReVOS, and Ref-Youtube-VOS, MoVISA achieves significant gains, notably a 13.2% J and F improvement on MeViS and an 8.4% J and F improvement on ReVOS.
By Ruining Zhao, Ho Kei Cheng, Alexander G Schwing
Passive LWIR hyperspectral ranging estimates distance in low-light scenes by exploiting atmospheric absorption in thermal radiance. The proposed TEDA method decouples range estimation from temperature–emissivity inversion, using a baseline estimator and transmittance extraction to recover atmospheric transmittance, then matching it to sensor-domain models for each candidate distance. TEDA reduces ranging bias, yields mean range estimates closer to LiDAR medians, and achieves a 20‑fold speedup over reference-range joint inversion.
By Zhihe Chen, Chen Fan, Shuo Liu, Xiaolin Huang, Yunze He, Xiaofeng He, Lilian Zhang
UpDown‑SC is a training‑free polar descriptor for indoor LiDAR place recognition that first canonicalizes gravity and then represents two complementary surfaces: the upper envelope of lower/middle structures and the lower envelope of overhead structures. By estimating a physical split from a cell‑balanced map height distribution and using a mask‑aware, non‑uniform two‑channel distance, it retains discriminative lower‑level evidence while limiting sensitivity to cross‑session variation. Experiments on repeated indoor sessions, mounting‑height changes, mixed outdoor‑to‑indoor trajectories, and an outdoor transfer sequence demonstrate more reliable first‑choice retrieval and significant gains over conventional Scan Context, while maintaining a lightweight CPU front end and supporting metric prior‑map localization.
By Jie Xu, Yongxin Yang, Ziyi Jin, Kangjin Yu, Hongjun Huang, Chao Han, Zhongpu Xia
The paper introduces a control‑free method for measuring large‑gradient deformation using spaceborne optical image correlation. By treating repeat‑pass correlation as a differential estimation problem, it jointly estimates surface displacement and revisit‑bias coefficients without surveyed ground control, providing a predictive accuracy envelope and per‑point uncertainty. Experiments on satellite data from recent earthquakes and glacier movement confirm the method’s low noise floor and robustness to stable‑terrain assumptions.
By Yueqiang Zhang, Chang Ma, Shuixin Pan, Haibo Liu
OREO is a framework that improves the visual fidelity of 3D generation models by using on-the-fly rendered and edited 2D views as pseudo-targets. It introduces a dynamic optimization loop where a 2D diffusion model refines rendered views, preserving geometry, viewpoint, and content while enhancing realism. These refined views serve as high‑quality supervision, enabling the 3D generator to learn from its own outputs and progressively improve its visual quality, outperforming pre‑trained baselines.
By Zhiyuan Ma, Wenbo Hu, Wang Zhao, Pengfei Wang, Ying Shan, Lei Zhang
The study investigates how the geometry of resting‑state functional connectivity (FC) influences classification of brain phenotypes and cross‑site transfer, rather than focusing on model design. It finds that FC variation across subjects is concentrated in a small effective subspace, and that differences in subspace orientation across cohorts can limit transfer even when effective ranks are similar. By projecting onto leading components at the effective‑rank scale, most classification performance is retained, and the alignment of site‑specific subspaces predicts transfer performance.
"whyItMatters":"The findings highlight that aligning effective subspaces across sites is crucial for improving generalization of FC‑based classifiers, offering a geometric diagnostic for cross‑site harmonization efforts."
By Xiao Fan, Jingyuan Li, Yubo Han, Hongbin Guo, Guanya Li, Yang Hu, Wenchao Zhang, Weibin Ji, Yi Zhang
The paper introduces Self‑Adaptive VLA, a post‑training method that lets Vision‑Language‑Action policies self‑adapt to deployment‑time hardware shifts by using rollouts as context. It creates shift‑conditioned expert demonstrations, compresses visual, proprioceptive, and action data into a latent context token, and modulates the policy via adaptive layer normalization. Experiments on four precision‑critical manipulation tasks show the method recovers over 80 % of the base policy’s performance under actuation bias and encoder offsets, and improves robustness on new workstations.
By Hongxin Zhang, Chunru Lin, Tsun-Hsuan Wang, Zhenjia Xu, Chuang Gan
The paper introduces OVIE, a monocular novel-view synthesis method that eliminates the need for multi‑view training data. By using a frozen depth estimator to generate pseudo‑target views from single images and applying masked and adversarial losses, OVIE is trained on 30 million uncurated images. It achieves state‑of‑the‑art performance on RealEstate10K and DL3DV, produces highly consistent multi‑view trajectories, and runs at 116 FPS—over 600× faster than the fastest baseline.
By Adrien Ramanana Rahary, Nicolas Dufour, Patrick Perez, David Picard
The paper presents a smartphone-based system that uses computer vision to automatically estimate vehicle speed and identify vehicles by license plate, make/model, and color. Experiments on a Brazilian dataset and real-world recordings in Austin, Texas show moderate recognition rates: 46% for license plates, 60.8% for color, 48.6% for make, and 16.89% for make/model. The study also discusses legal, technological, and practical considerations for deploying such smartphone recordings in traffic enforcement.
By Keya Li, Jahnavi Malagavalli, Lamha Goel, Tong Wang, Kara M. Kockelman
The paper introduces CARE, a lightweight, plug‑and‑play regularization framework for diffusion models that dynamically adjusts feature distributions based on condition similarity. By leveraging built‑in conditioning signals such as labels or text prompts, CARE promotes tighter feature clusters for similar conditions without requiring explicit alignment losses or external supervision. Empirical results show consistent improvements in visual fidelity and convergence stability, achieving significant FID reductions and speed‑ups on ImageNet and text‑to‑image tasks, and it can be combined with existing regularization methods for further gains.
By Fengjia Guo, Zhuoyi Yang, Jie Tang
The paper introduces TRACE, an amortized temporal gradient‑inversion attack that reconstructs private observation‑action trajectories from per‑step policy gradients in embodied reinforcement‑learning agents. TRACE exploits cross‑time correlation between gradients and exact action recovery from policy‑head gradients, achieving high reconstruction quality (18.8 dB PSNR) and near‑perfect action recovery with minimal computation. The study demonstrates TRACE’s effectiveness across various neural architectures and input modalities, and suggests that protecting temporal gradient streams may require sequence‑aware privacy mechanisms.
By Sudip Bhujel, Shanghao Shi, Ruiquan Huang, Ning Zhang, Yang Xiao
The paper introduces DEEPO, a Dual-Entropy Enhanced Policy Optimization method designed to mitigate hallucination in multimodal large language models (MLLMs). It addresses two weaknesses in reinforcement learning: (1) hard queries with high semantic entropy produce uniformly wrong samples, erasing advantage signals, and (2) confident-but-wrong tokens become invisible to gradients as the policy sharpens. DEEPO combines semantic‑entropy‑triggered expert prefixes to inject grounded continuations and Renyi preconditioning to counter logit saturation, yielding significant hallucination reduction while maintaining accuracy and training stability.
By Yingxuan Zhuang, Miao Pan, Wangjie Gan, Jingxiao Yang, Fan Wang, Weiming Liu, Cheng Tan, Xuhong Zhang, Jintao Chen
The paper introduces BaCVA, a Bayesian Context-aware personalized Value Alignment method for large language models. It treats personal values as priors and context-dependent preferences as posteriors, estimating contextual value salience from normative responses and using a dual-view personalization module to infer posterior preferences. Experiments show BaCVA outperforms strong baselines, offering more accurate and data‑efficient personalized value alignment.
By Hanze Guo, Aixuan Song, Jing Yao, Xiangxu Zhang, Xiaoyuan Yi, Xing Xie, Xiao Zhou
IndicBankBench is a 799‑case benchmark designed to evaluate the safety and reliability of language model assistants in Indian retail banking. It covers five operational domains, a capability/refusal domain, and twenty primary axes, assessing each case at four stages: safety, action and tool use, response adequacy, and advisory quality. The benchmark uses deterministic safety checks, a narrow resolver for ambiguous confirmation‑before‑write scenarios, and an LLM judge for semantic response adequacy, reporting strict pass rates that reveal a gap between strict reliability (43.7%–58.2%) and at‑least‑once success (60%–74%).
By Suvradip Paul, Chandra Bhushan, Harsh Sharma, Nitin Kukreja, Yatharth Dedhia, Keyur Doshi, Prashant Devadiga
SkinAgent AI is a multimodal, safety‑grounded framework designed for non‑diagnostic skincare support. It routes visual concerns (Acne, Pores, Wrinkles), estimates skin type from photographs, and provides count‑informed acne severity, all while grounding recommendations in a database and enforcing deterministic safety, privacy, and evidence checks. The system achieved high accuracy in routing and skin‑type estimation, and demonstrated no safety or privacy violations in controlled tests, though tool‑selection errors and incomplete product grounding remain.
By Muhammad Muhtasim Shahriar, Abdullah Mohammad Sayem, Tze Hui Liew, M. F. Mridha, Md. Mahiuddin
The paper investigates the existence of hallucination neurons in large language models by applying a rigorous five‑step diagnostic protocol to sparse probing methods. Using open‑source LLMs on TriviaQA, BioASQ, and NQ‑Open, the authors find that while detection of hallucination neurons replicates across models and datasets, the neurons are not uniquely localized—showing high feature correlation, moderate bootstrap stability, and weak overlap between sparse and dense rankings. The study demonstrates that sparse predictive structure can coexist with non‑unique neuron selection, underscoring the need for routine diagnostic validation in mechanistic interpretability.
By Huseyin Cavus, Sebin Sabu, Joshua Spear, Jaskaran Singh Kawatra, Pavithra Rajendran
PUBG Ally is an embodied, voice‑enabled AI teammate for PUBG: BATTLEGROUNDS that can perceive the game world, interpret player speech, and autonomously decide actions while keeping speech synchronized with gameplay. It combines a language‑model agent that uses a controlled interface to gather game information and a faster control layer for movement, combat, and recovery. The system was trained on nearly 39,000 real‑player sessions and evaluated through player feedback and preference comparisons, with live deployment requiring low‑latency on‑device execution and safety safeguards.
By Beomsoo Kim, Byeongju Kim, Dohyun Kim, Dongwon Kim, Eunchong Kim, Hongmin Kim, Hyeojung Im, Hyeonbin Hwang, Hyeonghwan Kim, Hyoseok Seol, Insub Im, Irene Chen, Jaeseung Jeon, Jimin Hong, Kiyoon Yoo, Minkyoung Park, Seohyeon Jung, Seungjun Chung, Sue Hyun Park, Sungwoo Kim, Youngin Cho, Yujeong Son, Kangwook Lee, Hyunseung Kim