Wayfarer is a domain‑agnostic, online deep RL agent that discovers options via Laplacian representation learning from high‑dimensional observations and uses them for control. The discovered options improve exploration, accelerate credit assignment, and generalise to unseen settings, leading to faster learning of complex policies. Wayfarer achieves state‑of‑the‑art performance among single‑stream agents on the most challenging Atari 2600 games, especially those requiring long‑horizon exploration such as Montezuma's Revenge and Private Eye.
By Erik M. Lintunen, Marlos C. Machado
SafeStreamingFlow is a goal‑conditioned planner that aligns flow sampling dynamics with execution dynamics by sequentially integrating a learned state vector field with hierarchical state prediction. It enforces safety constraints only for the executed step using high‑order control barrier functions, reducing planning latency and improving safety compared to prior safe diffusion/flow planners. The method demonstrates competitive goal‑reaching success across navigation, racing, and locomotion benchmarks.
By Seunghwan Jang, Jeongyong Yang, Siddharth Ancha, SooJean Han
The paper introduces Sigma, a large-scale continuous diffusion language model (3B/8B parameters) that uses steerable, low-dimensional ODE/SDE latent trajectories to address the non-smoothness of discrete diffusion models. Sigma is trained blockwise via likelihood optimization, jointly denoises Gaussian-corrupted token embeddings, and learns an optimal embedding geometry, leveraging pre-trained autoregressive weights for faster training. During inference, classifier-free guidance and score temperature are identified as essential for high-fidelity reasoning and coding, and Sigma matches or exceeds discrete models on benchmarks such as GSM8K, Minerva, HumanEval, MBPP, MATH-500, and AIME, while also revealing unique structural benefits like embedding-space steering and graceful degradation for low NFEs.
By Zhihan Yang, Wei Guo, Jean-Marie Lemercier, Simon Welker, Yonggan Fu, Mohammad Mahdi Kamani, Sajad Norouzi, Julius Berner, Tomas Geffner, Karsten Kreis, Yongxin Chen, Molei Tao, John Thickstun, Pavlo Molchanov, Ante Juki\'c, Arash Vahdat, Morteza Mardani
The paper introduces OMVV, an online multi-verifier algorithm that maintains a pool of weak verifiers with varying costs and performance. It adaptively selects a verifier each round using an online score combiner and exponential-weights routing, providing distribution-free guarantees on false-accept and false-reject rates. Experiments on reasoning benchmarks show OMVV achieves higher accuracy at lower verification cost than any single fixed verifier across different budgets.
By Erfan Hajihashemi, Yanning Shen
Pivot‑SD is an offline self‑distillation framework for masked diffusion language models that focuses training on high‑impact commitments, called pivots, identified by an information‑gain metric. By supervising only these pivots—using cross‑entropy for successful trajectories and targeted unlikelihood for failed ones—Pivot‑SD improves LLaDA‑8B‑Instruct on math and code benchmarks with just 200 questions and four rollouts each.
By Seo Hyun Kim, Sunwoo Hong, Younwoo Choi, Chen-Hao Chao, Se-Young Yun, Rahul G. Krishnan
arXiv:2610. 02217v1 Announce Type: cross Abstract: Large sparse linear systems from PDE discretizations require correction subspaces whose operator images explain the current residual.
By Zhongyan Ouyang, Weixin Liao, Mingquan Feng, Yehui Tang, Junchi Yan
Dynamic LLM routers aim to reduce inference costs by directing each query to the cheapest capable model. In a study of six commercial routers across 14 settings and eight task categories, none surpassed a simple random router that selects between two well-chosen models at the same cost, with some underperforming by over 10 percentage points. The authors identify four common patterns—difficulty blindness, length reversal, semantic matching, and roster suboptimality—that explain this gap and propose a new evaluation method and a simple two-model router that mitigates these patterns, though its advantage over random routing remains modest.
By Sam Wang, Julia White, Sahibzada Allahyar, Dhruv Atreja, Urchade Zaratiana, Kelton Zhang
The paper introduces Turnover-Orthogonal Credit Assignment (TOCA), a method for open-team multi-agent reinforcement learning that separates the effects of agent actions, pure turnover, and action–turnover interactions in the team return. TOCA uses a permutation‑invariant centralized critic with event tokens to provide counterfactual per‑agent credit signals and a softly weighted interaction variant, TOCA‑β, for high‑variance environments. Experiments show that TOCA outperforms event‑aware MAPPO critics and that removing interaction credit degrades performance, especially in high‑turnover settings such as the Dynamic Spread benchmark.
By Amit Thakur, Mukesh Singhal
The paper introduces a distributionally robust framework for survival analysis that simultaneously tackles latent subpopulation shift and outlier contamination. It employs an outer minimization to refine the nominal distribution by down-weighting contaminated samples and an inner maximization to target the most challenging subpopulation, directly handling non-decomposable survival losses such as the Cox partial log-likelihood. An alternating gradient-based algorithm, guided by KKT conditions, is developed, and experiments on simulated data and two benchmarks show improved worst-group performance and stable training under contamination.
By Seonghwi Kim, Sung Ho Jo, Minwoo Chae
The paper introduces FreqMo, a motion representation that splits human motion into wavelet frequency bands to handle varying temporal scales. By using Unified Frequency Residual Quantization, it compresses tokens threefold and supports stable single-stage generation. Experiments demonstrate state‑of‑the‑art fidelity, especially in preserving high‑frequency motion details, and the method is compatible with continuous diffusion models.
By Yunjiao Zhou, Junlang Qian, Gen Li, Xinying Guo, Lihua Xie, Jianfei Yang
The paper evaluates how robust large language models (LLMs) are when using retrieval‑augmented generation (RAG) in practical settings. It investigates whether RAG always outperforms non‑RAG approaches, whether adding more retrieved documents helps, and whether the order of documents matters, using a benchmark of 1,891 samples across five datasets and three task categories. Experiments with 11 LLMs show generally high retrieval robustness, but performance varies by task and prompting strategy, indicating that adopting RAG should be considered on a case‑by‑case basis.
By Shuyang Cao, Karthik Radhakrishnan, David Rosenberg, Steven Lu, Pengxiang Cheng, Lu Wang, Shiyue Zhang
FUSEye is a lightweight training framework that adapts a frozen COCO‑pretrained YOLO26‑x detector for fisheye images by adding only about 227k parameters. It introduces three modules: GridViews to enlarge compressed boundary regions, Z‑Adapters to correct distortion‑induced feature misalignment, and AgreeFusion to fuse detections across overlapping views. On the WoodScape fisheye benchmark, FUSEye boosts YOLO26‑x mAP50 from 0.148 to 0.266 while requiring only 25% of the labeled data to achieve 0.2597 mAP50, and it also improves YOLOv8‑11 detectors.
By Wenya Su, Kai Luo, Di Wen, Ruiping Liu, Yufan Chen, Junwei Zheng, Kunyu Peng, Kailun Yang
The paper introduces HypoDepth, an event-image monocular depth estimation framework that uses a discrete Depth Hypothesis Volume (DHV) to convert depth regression into a constrained search problem. By building a lightweight 3D cost volume between DHV features and contextual features, the method performs multi-scale correlation search for stable residual optimization, enabling efficient global-to-local refinement across resolutions. Experiments on DSEC and MVSEC show state‑of‑the‑art performance, strong zero‑shot generalization, and real‑time capability on resource‑limited devices.
By Daikun Liu, Teng Wang, Changyin Sun
The paper introduces threat‑preserving representation sensitivity (TPRS) to assess how changes in agent‑visible wording affect attack success rates (ASR) while keeping the underlying task and evaluation criteria constant. Experiments on Agent Security Bench, MCPTox, and AgentDojo show that renaming threat‑related tools can significantly alter ASR—up to +13.21 percentage points on some models—yet may also degrade benign utility. The findings suggest that security scores based on a single representation may not generalize, urging robustness claims to be validated across multiple threat‑preserving representations.
By Neeraj Karamchandani, Piyush Nagasubramaniam, Xinhong Xie, Sencun Zhu, Dinghao Wu
OLMo-Detect is a new benchmark for membership inference attacks (MIAs) on large language models, covering pre‑training, mid‑training, and post‑training stages. It aligns member and non‑member samples on key axes and rigorously filters non‑members using infini‑gram, also offering a shifted variant to test distribution robustness. Evaluation of 15 unsupervised and 3 supervised MIAs on the OLMo 2 family shows limited overall performance (best AUC 0.68), peak detection at mid‑training driven by data type, and sensitivity to distribution shifts, with findings generalizing to OLMo 3 and other models.
By Tao Shi, Chaoyi Xiang, Qiongkai Xu, Jey Han Lau
Gaze Attention is a new mechanism for multimodal large language models that selectively focuses on relevant visual regions during each generation step, rather than attending to all visual tokens. By grouping tokens into spatial regions and using learnable context tokens to retain global information, it reduces attention computation and visual key‑value entries by up to 90%. Experiments on 13 image and 6 video benchmarks show that Gaze Attention matches or outperforms dense‑attention baselines while using fewer visual resources.
By Junha Song, Byeongho Heo, Geonmo Gu, Jaegul Choo, Dongyoon Han, Sangdoo Yun
The paper investigates whether adding a short classification instruction after a user’s prompt improves the ability of activation probes to detect malicious inputs in large language models. Across 13 safety benchmarks and three open‑weight model families, a classification suffix consistently boosts out‑of‑distribution detection (up to ~4 AUC points) compared to no suffix, and the benefit transfers to multi‑position pooling probes used in production. The improvement stems from the classification format itself rather than the specific content of the instruction, though the optimal suffix varies with the model and readout type.
By Elad David, Max Fomin
The study investigates how different distribution shifts affect the evaluation of drug‑target affinity (DTA) models, using 718,800 drug‑protein pairs from ChEMBL and BindingDB. It compares a baseline Morgan‑fingerprint + protein‑CNN model with 12 architectures that combine four drug representations and three ESM‑2 interaction modes, finding that mean validation RMSE rises significantly under protein‑cluster and dual out‑of‑distribution (OOD) scenarios. Model rankings remain consistent across chemical shifts but diverge under protein and dual OOD, indicating that architecture selection depends on the specific extrapolation context rather than just average error or training‑set size.
Why it matters: The results suggest that DTA benchmarks should be designed to reflect the chemical and target shifts expected in real deployment, ensuring that chosen models perform reliably when applied to new chemical series or protein targets.
By Minjae Chung, Clara Li, Malar Paavai Muthukumaran, Shaunna Wang, Aniket Ramkrishnan Iyer, Shaun Qien Yeau Tan, Harinishree Sathu, Micky C. Nnamdi, J. Ben Tamo, Benoit Louis Marteau, May Dongmei Wang
RT‑SFT is a method for text style transfer that uses roundtrip translation through a pivot language to strip stylistic information from monolingual corpora, creating pseudo‑parallel data. This data is then used to LoRA‑finetune an instruction‑tuned large language model as a stylizer, allowing the model to rewrite sentences in a target style while preserving meaning. Experiments across four style domains show that RT‑SFT surpasses state‑of‑the‑art approaches, including few‑shot in‑context learning, and offers effective retrieval augmentation for expert style domains with strict terminology.
By Ruoxi Liu, Philipp Koehn
ReFract is a new benchmark that tests whether language model agents can act appropriately based on a user’s role, a capability called Perspective Awareness. It contains 150 expert‑validated entries derived from anonymized domain support queries, and uses Text World Models to simulate the agents’ operating environments and generate perspective‑aware action trajectories. Current state‑of‑the‑art LLMs solve at most 69% of the tasks, with over half of their trajectories attempting perspective‑violating actions.
By Hainiu Xu, V\'{i}tor N. Louren\c{c}o, Mohnish Dubey, Yunfei Bai, Yulan He, Caroline Catmur, Aline Paes, Marco Caserta, Akash Chandrayan, Luca D'Angelo