The paper investigates whether pixels alone can determine an image’s origin—human, AI class, or specific generator—under adversarial edits. It establishes a minimax limit: the best possible robust acceptance gap equals the minimum total‑variation distance between the target distribution and attacked source distributions, independent of verifier design. The study also shows that practical public verifiers can fail before reaching this theoretical ceiling, highlighting the need to evaluate both statistical limits and deployed verifier behavior separately.
By Kai Yao
G$^2$PTQ is a post‑training quantization framework that improves large language models by combining first‑ and second‑order information in a globally supervised, block‑wise optimization. It refreshes gradient and Hessian estimates before each Transformer block and uses a trust‑region scaling mechanism to stabilize gradient steps, preventing exploding weight updates. The method achieves better alignment with full‑precision models and outperforms state‑of‑the‑art baselines across various model families and bit‑widths.
By Ruikang Liu, Haoli Bai, Yuxuan Sun, Qian Zhang, Wenzheng Cai, Yanqi Hao, Feiyu Wang, Weidong Zhong, Zhuang Wang, Tong Yang, Xiangsheng Zhou
DepthEvidence is a 4B multimodal language model that integrates dense metric depth predictions into language generation. It employs a camera‑conditioned decoder to produce full‑resolution depth maps and a dense‑to‑language interface that converts these predictions into object‑aligned geometry tokens. The model is trained with geometric supervision and instruction tuning, and it sets new state‑of‑the‑art results on a Depth‑VQA benchmark and on instance‑level metric depth estimation across nine datasets.
By Jiangning Wei, Yuan Yao, Miaomiao Cui, Mingsheng Li, Humen Zhong, Shuai Bai, Zhibo Yang
SPADE (Serendipitous Pareto Distance Evaluation) is a new metric for recommender systems that simultaneously considers item similarity, popularity, and user relevance. It projects items into a two‑dimensional space and computes a user‑specific Pareto frontier of maximally popular and historically similar items, then averages the minimum Euclidean distance from this frontier for correctly recommended test‑set items. Experiments on five datasets and five baseline algorithms demonstrate that SPADE effectively discourages algorithms from exploiting accuracy‑only metrics and reliably isolates serendipitous discoveries.
By Tobias Vente, Maarten Peirsman, Noah Dani\"els, Hannu Toivonen, Bart Goethals
BAT-CLIP is a trimodal alignment framework that jointly aligns intracranial EEG (iEEG) neural embeddings to both pretrained audio and text anchors within a shared, frozen audio‑text manifold. Unlike existing CLIP‑style brain‑speech models that anchor neural activity to a single modality, BAT‑CLIP leverages both audio and text to preserve temporal structure and linguistic separability. On the naturalistic Podcast benchmark, BAT‑CLIP produces more robust representations than bimodal CLIP baselines and demonstrates the value of self‑supervised foundation models for CLIP training.
By Suhyun Kim, Jinmo Han, Danny Dongyeop Han, Ahhyun Lucy Lee, Jewoon Lee, Yonghyeon Gwon, Zach Paris, Chun Kee Chung, Saewoong Bahk, Nam Soo Kim, Seong Jae Hwang, Jiook Cha
The paper introduces DeformView, a wide‑baseline multi‑view dataset with pixel‑level annotations of geometric inconsistencies, and evaluates existing multi‑view consistency‑scoring methods, finding they transfer poorly to forensic localization tasks. To address this, the authors propose DEFECt3R, a lightweight learning‑based classifier that leverages cross‑view feature relationships and hard negative supervision to localize inconsistencies at the pixel level, achieving better performance and fewer false positives. Ablation studies confirm the importance of feature representations and correspondence quality for localization.
By Xander Staelens, Alb\'eric Loos, Bert Ramlot, Hannes Mareen, Peter Lambert, Glenn Van Wallendael
UniAR is a unified framework that improves autism spectrum disorder (ASD) recognition by using multi-granularity prompt learning and a large multimodal model to generate diagnostic descriptions at word, phrase, and sentence levels. It aligns these semantic representations with visual evidence through a Mixture-of-Experts-based Multi-Scale Alignment Module, enabling robust ASD detection across heterogeneous data types. Experiments on four brain MRI and facial expression benchmarks show that UniAR outperforms state‑of‑the‑art methods, achieving 75.9% accuracy on MRI and 91.6% on facial benchmarks, with gains of 1.5 and 1.2 percentage points respectively.
By Lei Xin, Zeheng Wang, Jiayin Zhu, Shihong Huang, Fanhu Zeng, Changjiang Jiang, Dengbo He, Yutao Yue, Zhenglun Kong
CG-HAF is a global‑local fusion framework for ordinal acne severity grading that explicitly combines holistic facial severity probabilities with structured lesion‑burden descriptors such as lesion count, detection confidence, and lesion area. The model uses a lightweight, interpretable classifier to produce the final grade, achieving statistically significant improvements over global‑evidence‑only baselines, especially for severe cases. Cross‑dataset testing reveals that strong performance within a dataset does not automatically transfer, largely due to mismatched grading criteria rather than detection failures.
By Muhammad Muhtasim Shahriar, Md. Naimur Asif Borno, Saad Aloteibi, Mohammad Ali Moni
DyMD introduces a Distribution Matching Distillation framework that adapts teacher supervision and critic fitting to preserve interaction dynamics in few-step video generation. By employing temporal affinity–conditioned re‑noise sampling and dynamics‑guided fake‑score tracking, DyMD balances motion recovery with visual quality. The method distills a 14B teacher into a 1.3B student that achieves significant gains on embodied‑video benchmarks and downstream action planning tasks.
By Haojun Xu, Jie Huang, Xin Lu, Mingchen Zhong, Zihao Fan, Linjiang Huang, Si Liu
The paper introduces Highlight-Then-Summarize (H2S), a two-step approach that first highlights question-relevant evidence in long documents and then condenses it into a compact, question-conditioned summary before generating an answer. The authors built the H2S-Dataset with 6,647 examples spanning 11 benchmark families, and developed H2S-RL to reward evidence selection and summary construction. Evaluated on the H2S-Bench suite, the H2S-14B model outperforms larger open-source models, achieving the highest overall score and maintaining strong performance even with a reduced output budget.
By Zhaoyuan Xia (Peking University, Baidu Inc), Qinghongbing Xie (Tsinghua University), Yung Xiang Hue (Tsinghua University), Jianguang Jiang (Baidu Inc), Gaofeng Lu (Baidu Inc), Zhenyu Jiao (Baidu Inc), Xing Yuan (Baidu Inc), Dai Dai (Baidu Inc), Tong Mo (Peking University), Long Zeng (Tsinghua University)
PriceBench is a diagnostic benchmark that extracts price, quality, and brand preferences from large language models (LLMs) by analyzing their hotel booking choices. Using a logit choice model, the study evaluated 28 LLMs from eight providers across 3,600 booking tasks involving 179 New York City hotels. Results show that more capable LLMs exhibit stronger, more consistent preferences, while weaker models either lock onto a single position or show near-indifference, with significant variation in price sensitivity and price/quality trade-offs across providers.
By Pavel Kireyev
The paper presents the first systematic evaluation of cultural awareness in large language models (LLMs) for Haitian Creole, a low‑resource language. Using a benchmark of culturally salient prompts curated by native speakers, the study assesses four dimensions—specificity, bias, diversity, and variation—in a text‑infilling setting. Results show a clear gap between Haitian Creole and higher‑resource French, with Haitian performance more uneven and more affected by French linguistic interference; story generation reveals stereotypical portrayals of Haitian characters even in positive contexts.
By Christelle Clervilsson, Yanzhu Guo
ClearGS is a method for 3D Gaussian Splatting from handheld videos that addresses uneven viewpoint coverage and mixed frame quality. It introduces Reliability‑aware View Allocation to assign graded supervision weights based on appearance reliability, degradation risk, and geometric utility, and weakly reactivates suppressed frames to maintain trajectory coverage. Additionally, Render‑Guided In‑Video Restoration uses a frozen no‑reference restoration expert and perceptual scoring to recover details lost to blur or distortion, followed by Full‑Trajectory Repair Consolidation to preserve early‑introduced details. On the GS2E and GSOTM datasets, ClearGS achieves state‑of‑the‑art performance with consistent CLIP‑IQA and MUSIQ gains and LPIPS reductions across most degradation settings, all without paired sharp supervision or matched clean references.
By Xuanzhi Liu, Xinyi Wu, Hang Pan, Wensi Huang, Zhenyao Wu, Ruize Han, Song Wang
OC-GS introduces an object‑centric Gaussian splatting method that refines the angle of each image while keeping a shared camera, rotation axis, and pivot for turntable reconstruction. By jointly optimizing image‑derived geometry and angles, it reconstructs objects from sparse, irregular captures and outperforms four pose‑free Gaussian splatting baselines across various view counts. Ablation studies confirm that both image‑derived angle initialization and the shared motion model are essential for the observed improvements, with real captures showing a 0.70 dB PSNR gain.
By Jae Joong Lee, Bedrich Benes
The paper explores whether natural‑language documentation aids coding agents in fixing software bugs and introduces a roundtrip benchmark that evaluates code descriptions by regenerating code and testing it. It finds that description completeness, not length, determines fidelity, and presents an optimizer that can produce fully faithful descriptions that generalize to new files. However, experiments across two model families and ten repositories show that such compact documentation does not improve an agent’s ability to resolve real repository issues compared to using the issue alone.
By Md Shohel Arman, Igor Molybog
The paper introduces Stepwise Marginal Information Gain (MIG), an intrinsic process reward that evaluates how each reasoning step of a large language model (LLM) or vision-language model (VLM) improves the likelihood of the reference answer. MIG rewards only new likelihood maxima, preventing duplicate credit, and is combined with outcome, format, and self‑distillation objectives to guide training. Experiments on eight benchmarks show that this method outperforms outcome‑only reinforcement learning and improves accuracy by up to 4.8 points over binary‑reward training, including a 12.6‑point gain on MathVerse and a 12.9‑point advantage on vision‑language transfer at 7B parameters.
By Xiangwei Wang, Wei Wang, Ken Chen, Nanduni Nimalsiri, Sachith Seneviratne, Saman Halgamuge
FlyAOC is a benchmark that tests AI agents on end‑to‑end ontology curation of Drosophila scientific literature. Given a gene symbol, a brief description, a large paper corpus, and ontology resources, agents must search for evidence and produce structured annotations such as function terms, expression patterns, and historical synonyms. The benchmark contains 7,397 expert‑curated annotations across 100 genes and evaluates different agent harnesses, revealing system‑level failure modes that single‑task evaluations miss.
By Xingjian Zhang, Sophia Moylan, Ziyang Xiong, Qiaozhu Mei, Yichen Luo, Jiaqi W. Ma
GT-HarmBench is a benchmark that evaluates AI safety risks in multi-agent settings, covering 1,535 high-stakes scenarios based on game-theoretic structures like the Prisoner's Dilemma, Stag Hunt, and Chicken. The benchmark draws scenarios from realistic AI risk contexts in the MIT AI Risk Repository and tests 15 frontier models, finding that agents fail to choose socially beneficial actions in 38% of cases, including military escalation, election manipulation, and medical malpractice. The study also measures how prompt framing and ordering affect outcomes and shows that game-theoretic interventions can improve socially beneficial outcomes by up to 18%.
By Pepijn Cobben, Xuanqiang Angelo Huang, Thao Amelia Pham, Isabel Dahlgren, Terry Jingchen Zhang, Zhijing Jin
Agentick is a unified benchmark for sequential decision‑making agents that evaluates RL, LLM, VLM, hybrid, and human agents on 37 procedurally generated tasks across six capability categories, four difficulty levels, and five observation modalities via a single Gymnasium‑compatible interface. It includes a Coding API, oracle reference policies, pre‑built SFT datasets, a composable agent harness, and a live leaderboard. An evaluation of 27 configurations and over 90,000 episodes shows no single approach dominates, with GPT‑5 mini leading overall, PPO excelling in planning and multi‑agent tasks, and the reasoning harness boosting LLM performance by 3‑10×, while ASCII observations outperform natural language.
By Roger Creus Castanyer, Pablo Samuel Castro, Glen Berseth
AcuityBench is a new benchmark that tests whether language models can correctly identify the urgency of medical care needed from user presentations. It unifies five public datasets—user conversations, online forum posts, clinical vignettes, and patient portal messages—under a shared four-level acuity framework, providing 914 cases for evaluation. The benchmark supports both explicit four-way classification and free-form conversational responses, revealing that models vary widely in accuracy and that conversational formats reduce over-triage but increase under-triage, especially for high-acuity cases.
By Robin Linzmayer (Department of Computer Science, Columbia University, Department of Biomedical Informatics, Columbia University), Georgianna Lin (Department of Biomedical Informatics, Columbia University), Di Coneybeare (Department of Emergency Medicine, Columbia University Irving Medical Center), Jason Chu (Department of Emergency Medicine, Columbia University Irving Medical Center), Trudi Cloyd (Department of Emergency Medicine, Columbia University Irving Medical Center), Manish Garg (Department of Emergency Medicine, Columbia University Irving Medical Center), Miles Gordon (Department of Emergency Medicine, Columbia University Irving Medical Center), Elizabeth Hartofilis (Department of Emergency Medicine, Columbia University Irving Medical Center), Benjamin Hong (Department of Emergency Medicine, Columbia University Irving Medical Center), Ashraf Hussain (Department of Emergency Medicine, Columbia University Irving Medical Center), Eugene Y. Kim (Department of Emergency Medicine, Columbia University Irving Medical Center), Oluchi Iheagwara King (Department of Emergency Medicine, Columbia University Irving Medical Center), Ross McCormack (Department of Emergency Medicine, Columbia University Irving Medical Center), Erica Olsen (Department of Emergency Medicine, Columbia University Irving Medical Center), John K. Riggins Jr (Department of Emergency Medicine, Columbia University Irving Medical Center), Mustafa N. Rasheed (Department of Emergency Medicine, Columbia University Irving Medical Center), Dana L. Sacco (Department of Emergency Medicine, Columbia University Irving Medical Center), Vinay Saggar (Department of Emergency Medicine, Columbia University Irving Medical Center), Osman R. Sayan (Department of Emergency Medicine, Columbia University Irving Medical Center), Amit Shembekar (Department of Emergency Medicine, Columbia University Irving Medical Center), Janice Shin-Kim (Department of Emergency Medicine, Columbia University Irving Medical Center), Wendy W. Sun (Department of Emergency Medicine, Columbia University Irving Medical Center), Bernard P. Chang (Department of Emergency Medicine, Columbia University Irving Medical Center), David Kessler (Department of Emergency Medicine, Columbia University Irving Medical Center), No\'emie Elhadad (Department of Computer Science, Columbia University, Department of Biomedical Informatics, Columbia University)