The paper introduces MOPC, a memory‑augmented online test‑time adaptation framework for EEG motor‑imagery BCIs that operates without source data. It uses random‑mask perturbations to enforce prediction consistency, a confidence‑calibrated reweighting loss to handle pseudolabels and class imbalance, and a lightweight memory module to prevent catastrophic forgetting. Evaluated on BCI IV IIb and OpenBMI datasets, MOPC outperforms prior methods by 2.22% and 3.67% while maintaining per‑trial adaptation latency of 40–60 ms.
By Kunkun Zhang, Kehao Wu, Zhitao Chen, Haigen Hu, Qianwei Zhou
Reverse target fishing seeks protein targets for a query molecule, a key step in drug discovery. The authors reframe this as a generative retrieval problem, introducing GenFish, which uses residual vector quantization to map protein language model features to shared semantic identifiers and autoregressively predicts protein codes from a molecular fingerprint. They also present ProTargetBench, a proteome‑scale benchmark covering 20,431 human proteins that distinguishes experimentally inactive from unmeasured associations, and show that GenFish outperforms both discriminative neural and non‑neural baselines on full‑catalogue retrieval metrics.
By Zhichao Deng, Haichuan Tan, Xin Hong, Huaizhi Wang, Tao Huang, Bowen Gao, Yanyan Lan
The paper introduces LLM‑Auction, a learning‑based generative auction mechanism that integrates ad allocation with large language model (LLM) generation. It redefines the auction object from discrete ad slots to distributions over LLM outputs, aligning allocation with advertiser value and user experience while inherently modeling allocation externalities without extra inference cost. The authors prove allocation monotonicity and continuity, show a first‑price payment rule has favorable incentive properties, and demonstrate state‑of‑the‑art allocation efficiency in a simulated LLM‑as‑a‑judge environment.
By Chujie Zhao, Qun Hu, Shiping Song, Dagui Chen, Han Zhu, Jian Xu, Bo Zheng
SurveyGraph introduces a framework that uses a Methods‑Entity Knowledge Graph (MEKG) to encode relationships such as follow‑on, competition, and shared‑resource links between research methods. The graph enables core‑paper discovery, hybrid outlining, and evidence‑gated narrative synthesis, outperforming flat lexical baselines in recall and achieving strong citation support and semantic edge validation. The study demonstrates MEKG’s effectiveness through controlled benchmarks, diagnostic tests, and human adjudication, positioning it as an auditable structural layer for survey generation.
By Yifan Yuan, Yancheng Zhu, Zuoyu Zhang, Xiaogang Peng
The paper introduces the Lightweight Frame Compensator (LFC), a 77K‑parameter module that can be inserted before any frozen segmentation backbone to improve real‑time polyp segmentation when inference is performed only on every Kth frame. LFC operates on a three‑frame window, fusing consecutive frames with spatio‑temporal attention into a single compensated frame, and is trained solely on pseudo‑labels generated by a differently seeded model, avoiding the need for expert annotation. Experiments on three polyp video benchmarks show that LFC restores most of the accuracy lost by frame skipping, sometimes matching frame‑by‑frame inference while delivering a 2.05×–2.55× increase in throughput, and that its benefit is greatest for high‑capacity Transformer backbones.
By Weihao Gao
HiP3R introduces a hierarchical pose graph approach for long‑horizon 3D reconstruction, addressing the trade‑off between chunk‑level and frame‑level optimization. It first builds a robust chunk‑level Sim(3) graph to set the global layout and scale, then fixes this layout while estimating a single SE(3) camera pose per frame to merge overlapping predictions into a unified trajectory. Experiments on multiple benchmarks show that HiP3R improves camera trajectories and 3D reconstruction quality for both chunk‑based and streaming backends without extra training.
By Zunnan Xu, Shangchen Zhou, Yihang Luo, Chen Change Loy
The paper introduces a new class of algorithms that embed existing automatic differentiation (AD) particle filter methods, enabling a bias‑variance tradeoff and reducing mean squared error in likelihood estimation for partially observed nonlinear stochastic dynamical systems. Unlike prior approaches, these algorithms only require a differentiable simulator of the latent dynamics, not explicit transition probabilities. Numerical experiments demonstrate that a hybrid method—using AD to refine a coarse solution from iterated filtering—outperforms current state‑of‑the‑art techniques on a challenging scientific benchmark.
By Kevin Tan, Aaron J. Abkemeier, Kunyang He, Giles Hooker, Edward L. Ionides
The paper introduces a continuous‑time linear‑quadratic model for agents learning to intercept a moving target, offering optimal control strategies that span reactive and predictive extremes. The model’s few interpretable parameters allow closed‑form solutions for optimal strategies, policy learning dynamics, and value learning dynamics. Analysis of the model’s eigenvalue spectra reveals a gap that explains why agents typically learn to reach the target before fully optimizing their movement.
By John J. Vastola, Kanaka Rajan
The paper introduces a wavenumber-aware deep learning approach that represents each Raman spectral point as an intensity–wavenumber pair and uses Transformer positional encodings derived from the wavenumber. This method remains robust to noise, sparse sampling, and partial spectral coverage across various neural network architectures, outperforming index-based and learned encodings when only part of the spectrum is observed. The approach was validated on Raman spectra from three bacterial species, their synthetic mixtures, and a public 30-isolate bacterial benchmark, demonstrating consistent performance on unseen sampling grids.
By Jihwan Kim, Jaeseong Heo, Yujin Kim, Sang Bin Yoon, Chanjoong Kim, Kwang-Im Oh, Hyon Bin Na, Kyungwon Kwak, Myeongkee Park, Joonyoung F. Joung
The paper introduces a partition theory of AI evaluation, arguing that an evaluator’s ability to certify a model’s capabilities is limited by its perception. It shows that if a model’s input partition refines the evaluator’s, it can pass all tests the evaluator can grade, but only a stricter evaluator can raise the discrimination floor. The authors define an echo test that is complete for the evaluator’s tests, proving that passing it only guarantees a minimum discrimination level and cannot detect exact imitation of weaker models.
By Ahmed M. Mahran
The paper introduces a new framework for evaluating sparse autoencoders (SAEs) by requiring their features to form an atomic interpretable unit basis, a coordinate system for semantically meaningful explanation. It formalizes this objective and organizes existing and new metrics, such as DAME, into a comprehensive benchmarking framework. Experiments across many SAE architectures reveal that some models appear monosemantic under current metrics yet are functionally polysemantic, with dominant features being less interpretable and more redundant.
By Riteng Zhang, Xiaoqian Wang
Verus-Insight is a new artifact detection framework for AI‑generated videos that treats detection as a structured process with three complementary capabilities: Inspection Planning, Evidence Verification, and Evidence Aggregation. The system is trained using human‑anchored full‑process supervision and progressive capability training, enabling it to perform fine‑grained artifact analysis and robust video‑level source prediction. A new benchmark, Verus‑Bench, provides fine‑grained human annotations of perceptible artifacts, and experiments show that Verus‑Insight outperforms existing methods in both artifact recovery and overall detection accuracy.
By Zhuoran Zhang, Yang Shi, Qixun Wang, Tengfei Liu, Yuqi Tang, Xuanyu Zhu, Yizhuo Jia, Jingyun Hua, Shixuan Liu, Xianghua Ying, Haotian Wang, Yuanxing Zhang, Pengfei Wan, Wenjing Yang
AgentVQA is an offline benchmark designed to evaluate vision‑language models on realistic agentic scenarios across five domains: Web Agents, Robotics, Egocentric Videos, Games, and Spatial Planning. It comprises 1,000 carefully selected real‑world examples drawn from 23 source datasets, enabling reliable comparisons without costly online simulations. The benchmark reveals that current models achieve only 58.5% accuracy, with most errors stemming from difficulties in grounding and tracking evolving visual states rather than from task misunderstanding or reasoning.
By Advait Gupta, Sachin Konan, Tom Yin, Gabriel Sarch, Zhuang Liu
The paper introduces OpenAttention, a sparse attention framework that uses a single count‑aware mean proxy to represent the unread tail of the key‑value cache, outperforming sampling methods at low read rates. It combines clustering, an ErrorGate that allocates reads based on estimated weight error, and a query‑geometry selector to efficiently compress keys while preserving query direction. Experiments on RULER‑32K HARD with Llama‑3.1‑8B show OpenAttention retains 99 % of dense accuracy at a 1 % read and achieves state‑of‑the‑art accuracy across models, contexts, and datasets, with a custom kernel that is ten times faster than dense attention at the same read rate.
By Hyunwoo Yu, Jongbeom Lee, Hyunbin Kim, Sungjun Park, Jaemin Choi, Jincheol Yang, Seunghun Moon, Bongjoon Hyun, Kyomin Sohn, Kyeongbo Kong, Suk-Ju Kang
The paper introduces Active Preference Acquisition (APA), a method for web agents to decide which unknown user preferences to ask about by evaluating the Prospective Personalization Value (PPV)—the expected cumulative benefit across future tasks minus the cost of asking. APA calculates counterfactual gains for each candidate preference using task prototypes and aggregates them by expected reuse, enabling agents to prioritize questions that may not help immediately but improve long‑term performance. Experiments on the ActiveWebPref benchmark show APA can acquire a large share of valuable preferences and achieve higher net utility than baseline strategies, demonstrating that considering future reuse is crucial for effective personalization.
By Yiqi Wang, Zizhi Zeng, Jiaqi Zhang, Yiqun Duan, Mingkai Zheng, Taotao Cai
AnyGameRenderer is a video‑to‑video generative renderer that transforms low‑quality, AI‑generated game footage—such as gray blockouts and colored primitives—into high‑quality visuals with detailed characters, motion, and effects while preserving layout, camera, and entity positions. It works across diverse game genres (tower defense, auto‑battlers, top‑down survivors, turn‑based battles, kart racing, crowd runners) and accepts plain video inputs without requiring complex engine data. Trained on ~10k paired clips from agent‑made games, Unreal Engine simulations, and real gameplay, it improves entity correspondence, raising temporal match from 50.9% to 62.9% and reducing missed count changes from 34.7% to 21.3%.
By Ruicong Liu, Liangyang Ouyang, Fengbo Lan, Zhixiang Wang, Kaipeng Zhang
The study evaluates wearable inertial measurement unit (IMU) systems for construction worker activity recognition using the VTT‑ConIoT benchmark. It systematically analyzes six classifiers, five sensor placements, and three window sizes under a leave‑one‑subject‑out protocol, revealing per‑subject accuracy ranging from 61.9 % to 88.7 % (mean 77.5 %). Findings show that multi‑sensor fusion and longer windows improve accuracy, standard augmentation offers modest gains, CTGAN augmentation degrades performance, and personalized calibration can recover up to 16.8 pp for the weakest participants.
By Ihsan Ul Haq (Centre for Computational Science and Mathematical Modelling, Coventry University, United Kingdom, School of Architecture and Built Environment Deakin University, Australia), James Brusey (Centre for Computational Science and Mathematical Modelling, Coventry University, United Kingdom), Farnad Nasirzadeh (School of Architecture and Built Environment Deakin University, Australia), Chandan Karmakar (School of Information Technology Deakin University, Australia)
ARCANA is a new benchmark designed to test far analogical reasoning, requiring mappings from real-world objects or phenomena to abstract grid-based transformations. Human participants reliably solve these tasks, whereas large language models exhibit a significant performance gap, highlighting systematic failure modes. The authors introduce ConceptSIR, a framework that improves LLM performance on challenging ARCANA tasks, yet far analogical reasoning remains an open problem for current models.
By Ruqi Chen, Zilei Wang, Wei Wang
The paper introduces a new neural network architecture that represents mixed‑monotone functions as the difference of two monotone subnetworks, enabling tight output enclosures via only two forward passes of opposite corners of an input interval. The enclosure size is a closed‑form function of the network weights, allowing direct minimization during training. Experiments on regression and dynamical‑system benchmarks show that the trained networks achieve enclosure sizes within 0.5% of a proven lower bound while maintaining prediction accuracy, and the method is significantly faster and often tighter than existing techniques such as CROWN, α‑CROWN, and zonotope propagation.
By Jonas Fingerhut, Lukas Koller, Matthias Althoff
The paper introduces Successive Scale Racing (SSR), a framework that improves agentic harness evolution by adaptively selecting optimal evidence combinations during each evolution round. SSR constructs diverse evidence sets based on task quality and coverage, then efficiently races candidate harnesses under a limited compute budget. Across three benchmarks, SSR outperforms baselines, steadily raising pass rates as the task pool expands from 40 to 200 while cutting validation executions by 60%.
By Jingzhe Liu, Jiliang Tang, Tong Zhao, Neil Shah, Clark Mingxuan Ju