Artificial Intelligence
RTSGameBench: Evaluating VLMs for Strategic Reasoning in Game Environments
San Kim, Daechul Ahn, Reokyoung Kim et al. (+3 more)
Vision-
Language Models (VLMs) often struggle with strategic
reasoning in a multi-
agent setting, i.e., deciding under uncertainty while coordinating with and adapting to other
agents. Real-time strategy (RTS) games can be a natural testbed for this capability, requiring long-horizon planning, coordination with allies, and adaptation to opponents under partial observability. But existing RTS
benchmarks cover limited
agent configurations, lack controlled evaluation across distinct strategic demands, and remain confined to fixed scenarios. To address these gaps, we propose RTSGAMEBENCH, built on Beyond All Reason, an RTS game whose larger scale broadens the strategic space beyond prior testbeds. In RTSGAMEBENCH, we offer holistic evaluation across diverse full-game matchups, diagnostic mini-games each emphasizing a focal strategic demand, and extensible coverage via a self-evolving framework that turns free-form queries into new mini-games, improving with accumulated experience. We further propose RTSGAMEAGENT, an
agent harness that combines scalable group-level trajectory planning with short- and long-term
memory and outperforms existing
agents. We find that state-of-the-art VLMs struggle where humans succeed, particularly when they must coordinate with allies or contend with multiple opponents.
Read the full abstract ▾
All authors San Kim, Daechul Ahn, Reokyoung Kim, Hyeonbeom Choi, Seungyeon Jwa, Jonghyun Choi
Artificial Intelligence
Decoupling Search from Reasoning: A Vendor-Agnostic Grounding Architecture for LLM Agents
Emmanuel Aboah Boateng, Kyle MacDonald, Amardeep Kumar et al. (+2 more)
Production LLM
agents increasingly depend on real-time search, yet native search grounding bundles
retrieval policy, provider choice, evidence injection, cost, latency, and generation behavior behind a single model-provider boundary. This coupling makes grounding hard to inspect, tune, reuse, or port, and can trigger Search-Induced Verbosity that breaks strict output contracts. We present Decoupled Search Grounding (DSG), a vendor-agnostic boundary that moves grounding outside the
reasoning model through an MCP-compatible gateway, exposing provider routing, source-aware context rendering, configured fallback,
retrieval-depth control, and exact plus semantic caching as first-class controls. Across five frontier models on SimpleQA, FreshQA, and HotpotQA, native search leads on recency-sensitive FreshQA, but DSG exposes a stronger frontier when control matters: on SimpleQA it nearly matches native accuracy (86.1% vs. 87.7%) at 91% lower search cost, preserves concise answer contracts, and reaches a 99.4% warm-cache hit rate with 68% lower latency. Deployed as a shared production grounding layer for large-scale agentic workloads with interchangeable models, DSG matches native-search accuracy on an e-commerce query-understanding (QIU) workload while cutting search cost by over 98%. Real-time grounding is best treated as an optimizable interface boundary, not a fixed model feature.
Read the full abstract ▾
All authors Emmanuel Aboah Boateng, Kyle MacDonald, Amardeep Kumar, Siddharth Kodwani, Sudeep Das
Artificial Intelligence
SciRisk-Bench: A Risk-Dimension-Aware Benchmark for AI4Science Safety
Linghao Feng, Yinqian Sun, Dongqi Liang et al. (+8 more)
Large
language models (LLMs) are increasingly embedded in AI for Science (AI4Science) workflows, from scientific question answering and literature analysis to laboratory planning and autonomous discovery. This progress creates an urgent need for safety
benchmarks that evaluate not only scientific competence, but also whether models recognize and avoid risks in high-stakes scientific contexts. Existing AI4Science safety
datasets cover several disciplines and task formats, leaving the underlying risk dimensions underspecified. We introduce SciRisk-Bench, a
benchmark designed to evaluate AI4Science safety from two complementary perspectives: explicit risk dimensions and scientific disciplines. SciRisk-Bench covers 7 disciplines, 31 subdisciplines and 10 risk dimensions. In the experimental section, we evaluate both mainstream LLMs and science-oriented LLMs across risk dimensions, disciplines, and sub-disciplines, enabling fine-grained diagnosis of where scientific models remain unsafe.
Read the full abstract ▾
All authors Linghao Feng, Yinqian Sun, Dongqi Liang, Sicheng Shen, Chenfei Yan, Yuxuan Peng, Yilin Zhao, Haibo Tong, Kai Li, FeiFei Zhao, Yi Zeng
Artificial Intelligence · Cross-listed from astro-ph.EP
TransitNet: A Compact Attention-Augmented Deep Learning Framework for Low-SNR Transit Blind Searches
Xingchen Yan, Jian Ge, Qingtian Liu et al. (+3 more)
Motivated by the observational incompleteness of intermediate-to-long-period Earth-size
planets, we present TransitNet, a compact
attention-augmented deep-learning framework for low-SNR transit blind searches. To enable realistic method development and objective threshold calibration under blind-search conditions, we develop a unified
dataset construction, benchmarking, and threshold-selection framework. On recovery
benchmarks constructed from unseen Kepler targets, TransitNet attains 95.2 percent accuracy in the challenging SNR range of 6 to 8 and outperforms both TLS and BLS, achieving ROC-AUC and PR-AP values of 0.974 and 0.982, respectively. In an injected Earth-size and sub-Earth-size transit recovery experiment, TransitNet achieves a recovery rate of 93.0 percent, substantially exceeding those of TLS (63.1 percent) and BLS (60.0 percent). In addition to detection, TransitNet provides
attention-based estimates of transit windows and midpoints. On an independent evaluation set, 97.4 percent of injected transits are fully covered by the estimated transit window. Applied to real Kepler observations, the model successfully recovers all 34 selected confirmed Kepler
planets, with a mean absolute transit midpoint error of 1.24 hours. The model combines a compact footprint of about 1.5 MB with high inference efficiency, yielding speed-ups of about 12 to 25 times relative to CPU-TLS and about 4 to 5 times relative to CPU-BLS. These results demonstrate that TransitNet provides an accurate, scalable, and computationally efficient framework for low-SNR transit blind searches in the tested regime and motivate its extension to longer-period Earth-size
planet searches.
Read the full abstract ▾
All authors Xingchen Yan, Jian Ge, Qingtian Liu, Kevin Willis, Quanquan Hu, Jiapeng Zhu
Artificial Intelligence · Cross-listed from cs.CL
As Easy as Rocket Science: Assessing the Ability of Large Language Models to Interpret Negation in Figurative Language
Jasmine Owers, Edwin Simpson and Martha Lewis
Figurative
language and negation are two areas that challenge current
language models, however, both are widely used throughout written and spoken
language. Large
language models (LLMs) are also widely used in everyday contexts where they cannot necessarily be tuned for a specific
dataset. It is therefore essential to understand the ability of LLMs to correctly interpret text that includes both negation and figurative
language. To investigate this, we develop a set of new annotations to an existing
dataset of figurative
language, and test a range of
language models on the
dataset. We find that the combination of negation and figurativeness can present a particular challenge, and that performance overall and across different negation types is particularly dependent on the
prompt style used.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.IR
SAERec: Constructing Fine-grained Interpretable Intents Priors via Sparse Autoencoders for Recommendation
Jiangnan Xia, Xuansheng Wu, Yu Yang et al. (+2 more)
Intent-based recommender systems have gained significant
attention for improving accuracy and interpretability by modeling the underlying motivations behind user behaviors. Most existing models derive intents directly from user sequences via clustering or prototype learning. However, they are sensitive to sequence quality, require presetting the number of intents, and lack explicit semantic grounding. These issues lead to an incomplete and coarse intent set and limit the effectiveness of recommendation. In this paper, we propose the Sparse Autoencoder for intent-based recommendation (SAERec), a novel recommender that automatically constructs a fine-grained and interpretable intent space from a textual corpus to guide recommendation. Rather than treating texts as side signals, SAERec leverages them as high information density evidence for intent construction. Specifically, we first extract a comprehensive set of fine-grained interpretable intents from the latent space of large
language models (LLMs) by using a sparse autoencoder (SAE) to disentangle and interpret text
embeddings, which isolates intent-related semantics from textual noise. Then, for each user, we retrieve relevant intents from this set as priors to guide recommendation. It contains personal intents matching a user's current interests and public intents capturing general item patterns shared across users (e.g., quality, price). Finally, to integrate retrieved intents into sequence modeling, we propose a multi-branch
attention mechanism that captures temporal dependencies and injects both personal and public intent signals, followed by an adaptive fusion layer to construct the final user representation for recommendation. Extensive experiments on public
datasets demonstrate the superiority of SAERec, consistently outperforming state-of-the-art baselines while providing human-understandable explanations.
Read the full abstract ▾
All authors Jiangnan Xia, Xuansheng Wu, Yu Yang, Xin Wang, Ninghao Liu
Artificial Intelligence
Zhimin Fan, Hongwei Yu, Yeqing Shen et al. (+9 more)
Improving GUI
agents typically relies on behavior cloning on expert trajectories. However, as the current policy deviates from the expert policy, it inevitably encounters policy-induced off-trajectory states during closed-loop execution, i.e., states that fall outside the expert trajectories. Since expert trajectories provide no demonstrations for these unseen states, such states receive no effective supervision, leaving the policy unable to select the correct action. To close this supervision gap, we propose Skill-Guided Continuation
Distillation (SGCD), an iterative self-improvement framework. SGCD first runs the plain policy without skill guidance for a few steps to reach realistic off-trajectory states. From these states, a skill-guided policy then completes the task and produces successful continuations, which are mixed with expert trajectories to supply supervision over policy-induced off-trajectory states. The skills are extracted from both successful and failed rollouts, consisting of Continuation Plans, Critical Targets, Failure Traps, and Success Criteria. On OSWorld-Verified, SGCD improves the success rate of three base models from the low-30% range to over 50%, demonstrating its effectiveness and generality.
Read the full abstract ▾
All authors Zhimin Fan, Hongwei Yu, Yeqing Shen, Haolong Yan, Guozhen Peng, Tianhao Peng, Yudong Zhang, Xiaowen Zhang, Kaijun Tan, Zheng Ge, Xiangyu Zhang, Daxin Jiang
Artificial Intelligence
Generative-Model Predictive Planning for Navigation in Partially Observable Environments
Thomas Quilter, Yifan Zhu, Guorui Quan et al. (+2 more)
Navigation in partially observable environments presents a significant challenge for autonomous
agents, requiring effective decision-making with limited sensory information in unknown environments. Belief-based methods, particularly those using neural networks to approximate the belief space, often fail to capture the inherent multimodality of belief spaces, especially in high-dimensional cases with perceptual aliasing. While generative models present a compelling alternative, they typically require substantial data or expert demonstrations and lack explicit mechanisms for long-term planning. In this paper, we introduce BeliefDiffusion, a novel framework that combines the benefits of both generation and planning. BeliefDiffusion leverages
diffusion models to explicitly characterize
multimodal belief distributions and utilizes Model Predictive Control (MPC) to simultaneously plan ahead. It consists of two steps: (1) Imagining plausible environment configurations based on observation history and (2) Planning efficient
navigation strategies across an aggregated configurations. Through extensive experiments in synthetic map environments, we demonstrate that BeliefDiffusion significantly outperforms both model-free reinforcement learning baselines and other generative approaches in
navigation success rate and path efficiency. Our results validate that explicitly incorporating
multimodal belief representations into planning enables more robust
navigation in partially observable settings.
Read the full abstract ▾
All authors Thomas Quilter, Yifan Zhu, Guorui Quan, Mingfei Sun, Samuel Kaski
Artificial Intelligence · Cross-listed from cs.LG
Domain-Shift Aware Neural Networks for Unbalance Characterization in Rotating Systems
Bernardo Feijó Junqueira, Claudio Kiyoshi Umezu, Bruno Bilhar Karaziack et al. (+2 more)
This work investigates the application of a domain-shift aware neural network for regression tasks aimed at estimating unbalance masses in rotating shafts under varying operating conditions. Experimental data were collected from a test rig in which a primary shaft, equipped with a flange carrying unbalanced masses, was driven at different rotational speeds, while a secondary shaft could be optionally activated to introduce domain discrepancy. The unbalance masses were positioned at a fixed radial distance, and the dynamic response of the system was recorded using triaxial accelerometers. The inverse problem of mass estimation is formulated within a domain adaptation framework, where the network is trained with a maximum mean discrepancy strategy to align feature representations across source and target distributions. The results demonstrate the effectiveness of explicitly addressing domain shift in improving prediction accuracy, especially when the system's physical behavior and sources of domain discrepancy are not fully known and fall outside the training conditions. These findings highlight the potential of domain-shift aware models for regression tasks in Structural Health Monitoring.
Read the full abstract ▾
All authors Bernardo Feijó Junqueira, Claudio Kiyoshi Umezu, Bruno Bilhar Karaziack, Tomaz Junior, Daniel Alves Castello
Artificial Intelligence
Making AI Scientists Auditable from Evidence to Claim
Zijian Wang, Hanqi Li, Ziyue Yang et al. (+17 more)
AI research systems can report improved results without establishing that the tested implementation matches the method described in their conclusions. We introduce Xcientist, a research harness that connects literature review, component-based idea generation, experiment validation and reporting. An evidence graph retains source passages, methods, baselines and evaluation conditions; component records track design revisions; and validation contracts specify the implementation, comparisons and artifacts required at each stage. We define claim drift as an unresolved mismatch between an accepted claim and its attributed sources, code or experiments. A retrospective audit across
agent memory, traffic
forecasting and physics-informed learning found audit claim drift rates of 3.6–16.7% for Xcientist, lower than those of each of three comparison systems in all nine task-by-relation comparisons. This measure includes both confirmed mismatches and unverifiable claims; most of the difference arose from fewer unverifiable records. The case studies also document performance gains, unsuccessful revisions and limits on component attribution. This technical report describes the system, its implementation and the records needed to check how a proposed idea became a tested method and a final claim.
Read the full abstract ▾
All authors Zijian Wang, Hanqi Li, Ziyue Yang, Zijian Hu, Shenghan Zuo, Yunzhe Zhang, Da Ma, Danyu Luo, Chenrun Wang, Jing Peng, Tiancheng Huang, Sijia Guo, Huayang Wang, Zichen Zhu, Senyu Han, Yilu Cao, Bo Chen, Xin Chen, Kai Yu, Lu Chen
Artificial Intelligence · Cross-listed from cs.LG
Scaling Learning-based AEB with Massive Unlabeled Data
Xiangyu Wang, Yang Zhan, Mengxiang Hao et al. (+9 more)
This paper studies how to scale learning-based automatic emergency braking (AEB) with massive unlabeled fleet data under production constraints. Our approach is based on meta-feedback semi-supervised learning (MF-SSL), where a teacher generates pseudo labels for unlabeled driving data and is updated using a small labeled anchor set as safety-critical feedback. In production, anchor ambiguity and labeled-unlabeled mismatch can amplify systematic pseudo-label errors, leading to spurious triggers. We propose a stabilized MF-SSL framework with (i) Noise-Aware Decoupling, which removes ambiguity-prone anchors from the teacher's supervised update path, and (ii) kinematics-gated pseudo-labeling with a teacher conflict penalty to suppress mismatch-induced risk
hallucinations on unlabeled data while maintaining broad coverage. Extensive experiments show consistent gains as unlabeled data scale from 1M to 1B windows, improving safety while keeping comfort stable. The 1B-trained student model is deployed to hundreds of thousands of vehicles and validated over \$10^9$ km of driving, achieving a positive-to-false activation ratio exceeding 100:1 and a 35% improvement in accident-free driving mileage over a production rule-only baseline.
Read the full abstract ▾
All authors Xiangyu Wang, Yang Zhan, Mengxiang Hao, Chuanchuan Zhong, Yansong Jia, Junjie Zhang, Yu Han, Xin Jiang, Zhen Cao, Ying Wang, Yulun Song, Zhitao Xu
Artificial Intelligence · Cross-listed from cs.CV
URDF Synthesis from RGB-D Sequences via Differentiable Joint Inference and Energy-Consistent Verification
Xinze Zhang
Reconstructing simulation-ready digital twins of articulated objects from sensor observations remains constrained by two persistent gaps: (i) part-level
geometric reconstruction is decoupled from kinematic-parameter estimation, and (ii) the recovered models often violate basic dynamic invariants such as energy conservation, leading to drift when the URDF is replayed in physics simulators. We present KinemaForge, a constraint-driven pipeline that jointly infers part-level shape, joint
topology, and joint parameters from short RGB-D sequences and validates the result against an energy-consistent verifier built on differentiable rigid-body dynamics. The pipeline introduces three components: a kinematic constraint graph that encodes joint-part incidences as soft edges; a differentiable screw-axis solver that backpropagates from rendered observations through Featherstone's articulated-body
algorithm to joint parameters; and an energy residual loss that penalises non-physical free responses of the reconstructed model. Across five PartNet-Mobility categories and an internal RGB-D
benchmark, KinemaForge reduces the average joint-axis error from 4.52 degrees to 2.83 degrees (-37.4%) over the strongest
geometric baseline (PARIS) and from 5.30 degrees to 2.83 degrees (-46.6%) over the interaction-based Ditto baseline, lowers long-horizon simulation drift by 64% (vs. PARIS) over 50 s rollouts, and yields URDFs whose closed-loop manipulation success rate improves by 14.6 percentage points over Ditto in our preliminary evaluation. Code and reconstruction data will be released upon acceptance.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CL
Aligning Implied Statements for Implicit Hate Speech Generalizability with Context-Bounded Semi-hard Negative Mining
Wicaksono Leksono Muhamad and Yunita Sari
Classifying implicit hate
speech remains a challenge, as intent is often masked through insinuation and context rather than explicit slurs. Prior supervised contrastive approaches improve in-domain detection but can overfit surface cues and struggle to transfer across
datasets. We propose ImpSH, a triplet-based framework that aligns posts with implied statements when available and uses context-bounded semi-hard negatives to focus learning on near confusions. We also examine AugSH, which forms positives via data augmentation. In controlled evaluations on IHC, SBIC, and DynaHate with BERT and HateBERT, ImpSH is a viable alternative to standard supervised contrastive baselines and often improves cross-domain performance under matched preprocessing and tuning budgets. Representation analysis using
alignment and uniformity indicates tighter positive pairs with balanced global spread, and qualitative nearest-neighbor case studies illustrate typical false negatives under domain shift. These results demonstrate that aligning posts with their implied statements via context-bounded mining provides a more stable, bijective-like mapping to related insinuations, overcoming the
volatility inherent in traditional clustering-based representation learning.
Read the full abstract ▾
Artificial Intelligence
WorldLines: Benchmarking and Modeling Long-Horizon Stateful Embodied Agents
Yehang Zhang, Jianchong Su, Haojian Huang et al. (+7 more)
To assist humans over extended periods in real homes,
embodied agents must remember user routines, world states, and past interactions. Existing long-term
memory benchmarks mainly evaluate
language-centric
retrieval and question answering, while
embodied benchmarks often focus on short-horizon task execution without testing long-term
memory use in dynamic environments. We introduce WorldLines, a project-driven
benchmark for long-horizon
embodied household assistance. It constructs temporally extended household traces with dialogues, actions, execution feedback, object and device state changes, and converts them into evidence-linked samples for
Memory QA and
Embodied Task Planning. We further propose ObsMem, an observer-grounded
memory framework that maintains visibility-aware
memories and action-native state trails for state-aware decisions. Experiments reveal persistent challenges in partial observability, overwritten world states, and translating long-term
memory into
embodied plans, while ObsMem offers a stronger reference architecture for this setting.
Read the full abstract ▾
All authors Yehang Zhang, Jianchong Su, Haojian Huang, Yifan Chang, Tianhao Zhou, Xinli Xu, Yingjie Xu, Yinchuan Li, Zexi Li, Ying-Cong Chen
Artificial Intelligence · Cross-listed from cs.MA
Skill-MAS: Evolving Meta-Skill for Automatic Multi-Agent Systems
Hehai Lin, Qi Yang and Chengwei Qin
Large
Language Model (LLM)-based automatic Multi-
Agent Systems (MAS) generation has become a crucial frontier for tackling complex tasks. However, existing methods face a dilemma between model capability and experience retention. Inference-time MAS leverages frozen frontier LLMs but repeats identical searches without learning from past experience. Conversely, Training-time MAS internalizes experience via gradient updates but is constrained by the low capability ceiling of smaller models, and is hard to scale to large frontier LLMs. To bridge this gap, we propose Skill-MAS, a novel third path that decouples experience retention from parametric updates by conceptualizing the high-level orchestration capability as an evolvable Meta-Skill. Skill-MAS refines this architectural knowledge through a closed optimization loop: (1) Multi-Trajectory Rollout samples a behavioral distribution for each task under the current Meta-Skill; and (2) Selective Reflection adaptively selects priority tasks and applies hierarchical contrastive analysis to distill systemic experience into generalizable, strategy-level principles. Extensive experiments across four complex
benchmarks and four distinct LLMs demonstrate that Skill-MAS not only achieves remarkable performance gains but also maintains a favorable cost-performance trade-off. Further analysis reveals that the evolved Meta-Skills are highly robust and exhibit strong transferability across unseen tasks and different LLMs.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.HC
Improving Human-Robot Teamwork in Urban Search and Rescue Through Episodic Memory of Prior Collaboration
Taewoon Kim, Emma van Zoelen and Mark Neerincx
Effective human-
robot teamwork requires
robots to adapt to partners, situations, and task dynamics from the start of an interaction. In the MATRX Urban Search and Rescue (USAR) environment, people can externalize collaboration patterns (CPs) they discover during teamwork through a chat and reflection interface. We study whether a
robot can use such prior team experience to become a better teammate in future interactions. To this end, we represent historical CPs as knowledge-graph episodic
memories and use graph representation learning with a node-classification objective to identify a representative and effective
memory for reuse. We then initialize the
robot with this
memory before a new collaboration episode begins. Across 20 participants and 160 round-level observations, initializing the
robot with a single automatically selected prior CP increases rescue success from 25.7% to 41.3% and reduces average task time by 283 seconds. The strongest gains appear at the beginning of interaction, suggesting that reusable episodic
memory can help
robots enter collaboration with more effective task knowledge and support smoother early teamwork.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Target-confidence Recourse Using tSeTlin machines: TRUST
K. Darshana Abeyrathna, Sara El Mekkaoui, Nils Enric Canut Taugbøl et al. (+1 more)
Counterfactual explanations are widely used to provide algorithmic recourse in high-stakes decision-making systems. Most existing methods seek the smallest change to an input that flips a model's decision. However, decision-makers often rely not only on predicted labels but also on confidence thresholds and risk margins. Counterfactuals that barely cross a decision boundary can be fragile and unstable under noise or model variation. In this paper, we propose Target-confidence Recourse Using tSeTlin machines (TRUST), a framework in which users explicitly specify the desired prediction confidence for recourse. Rather than generating counterfactuals and evaluating confidence afterward, TRUST directly searches for minimal changes that satisfy a user-defined confidence target, enabling comparison of recourse options in terms of cost, confidence, and robustness. We instantiate TRUST using a Probabilistic Tsetlin Machine (PTM) combined with
Bayesian optimization. The probabilistic clause-based structure of PTM links prediction confidence to the stability of decision rules. We show that counterfactuals satisfying the same rules can still differ substantially in reliability depending on how securely they satisfy those rules, revealing whether decisions are supported by robust or fragile clause activations. Experiments on synthetic and real-world
datasets demonstrate that target-confidence counterfactuals produce more robust and interpretable recourse than conventional boundary-based approaches. Across multiple
benchmarks, TRUST achieves perfect robustness while maintaining low recourse cost, including an L2 distance of 0.10 on the Haberman
dataset at 0.92 confidence. By explicitly controlling confidence and exposing rule-level stability, TRUST provides actionable recourse for high-stakes decision support.
Read the full abstract ▾
All authors K. Darshana Abeyrathna, Sara El Mekkaoui, Nils Enric Canut Taugbøl, Anuja Vats
Artificial Intelligence · Cross-listed from cs.CL
Beyond Reward Engineering: A Data Recipe for Long-Context Reinforcement Learning
Xiaoyue Xu, Sikui Zhang, Xiaorong Wang et al. (+2 more)
Long-context
reasoning is an essential capability for large
language models, particularly when they are deployed as autonomous
agents that must reason over lengthy trajectories. Reinforcement learning (RL) has recently emerged as a dominant paradigm for improving this ability, yet existing work largely focuses on reward engineering while diverse training data remains scarce. We revisit this problem from a data-centric perspective and show that a simple yet effective data recipe alone, paired with a minimal outcome-based GRPO setup, suffices to substantially improve long-context
reasoning. Our recipe targets three complementary task families –
retrieval, multi-evidence synthesis, and
reasoning – for which we construct and curate eight
datasets totaling ~14K examples. Experiments on three models (Qwen3-4B/8B/30B-A3B) yield average gains of +7.2/+3.2/+6.4 points across seven long-context
benchmarks, surpassing prior RL training sets. We further demonstrate that these gains transfer to agentic tasks, where continuing RL training on an
agent-tuned model with our data recipe improves
GAIA by +4.8 and BrowseComp by +7.0 points. We will release our
datasets to facilitate future research.
Read the full abstract ▾
All authors Xiaoyue Xu, Sikui Zhang, Xiaorong Wang, Xu Han, Chaojun Xiao
Artificial Intelligence · Cross-listed from cs.RO
Space Is Intelligence: Neural Semigroup Superposition for Riemannian Metric Generation
Chenghao Xu
Traditional approaches place intelligence in the
agent, whether as a learned policy or a search procedure. We instead place intelligence in the space itself: a scene induces a Riemannian metric on the configuration
manifold, and action reduces to following the geodesics of that metric rather than invoking a separate planner or collision checker. A single Encoder-Router network realizes this idea through three complementary parameter groups – frame parameters that orient the generators, modulation parameters that govern their spatial propagation, and basic coefficients that determine their strength. These groups combine through a shared
semigroup-superposition mechanism to produce a single Riemannian metric field, yielding a compact architecture whose geometry scales naturally with scene complexity. Trained on a single two-obstacle scene, the model demonstrates robust zero-shot generalization across unseen obstacle configurations, with orders-of-magnitude separation between collision-free and obstacle-penetrating path costs.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
When to Commit and When to Defer: Maturing Markov Decision Processes under Refining Information and Expiring Opportunities
Jiaxi Liu, Aiping Yang, Yuhang Yang et al. (+4 more)
Sequential decisions often become better informed as action opportunities expire. We introduce Maturing Markov Decision Processes (MMDPs), a structured subclass of augmented finite-horizon Markov decision processes, for organizing when to commit and when to defer. The Expiring-Actions-First Principle provides a sufficient condition for deferral based on information gain and delay loss. This structure motivates stage-local policies, action abstraction, and policy-guided planning. Replenishment experiments show improved learning and test performance under the MMDP interface and, in the primary setting, a larger test-performance gain from forecast refinement. In synthetic cash management, MMDP improves PPO test returns across both network sizes; five-account equal-size mask controls show that action count alone does not explain its learned gain, and the advantage persists with action abstraction and policy-guided search. A production-scale cash-management case study adds complementary evidence. Together, these results support commitment timing as a useful inductive bias for learning and planning.
Read the full abstract ▾
All authors Jiaxi Liu, Aiping Yang, Yuhang Yang, Shuqi Zhang, Zewei Dong, Jiangming Yang, Xuebin Chen
Artificial Intelligence · Cross-listed from cs.HC
SwitchBraidNet: Quantisation-Aware Lightweight Architecture for Hybrid Brain-Computer Interface
Gourav Siddhad and Yogesh Kumar Meena
Hybrid
brain-computer interfaces (BCIs) that integrate motor imagery (MI) and steady-state visual evoked potentials (SSVEP) provide high-dimensional neural decoding but typically exceed the computational limits of embedded hardware. To address this, we propose SwitchBraidNet, a compact
EEG classification architecture designed for low-power deployment. The model employs a dual-path temporal braid to extract multiscale oscillatory features, an adaptive squeeze-and-excitation spatial switch for electrode gating, and a log-variance readout layer for direct band-power encoding. Furthermore, through systematic quantisation-aware training on the OpenBMI
dataset, we compared SwitchBraidNet against four established baselines across FP32, FP16, and INT8 precisions. Experimental results demonstrate superior efficiency and performance, achieving MI accuracy of 69.49% (FP16), SSVEP accuracy of 93.48% (FP32), and a hybrid information transfer rate of 64.82 bits/min (FP16). With an INT8 footprint of only 3.03 KB, SwitchBraidNet maintains high accuracy across varying numerical precisions, demonstrating its suitability for low-power embedded BCI deployment.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Reinforcement Learning Foundation Models Should Already Be A Thing
Abdelrahman Zighem and Jill-Jênn Vie
Foundation models for
language and
vision are powered by internet-scale data, while structured domains such as tabular prediction are powered by synthetic data. This substitute shifts the challenge from collection to prior design. Such priors already exist for many structured tasks: TabPFN and its successors solve tabular classification with a
transformer pretrained on a synthetic
Bayesian prior. We make two points. First, reinforcement learning is the conspicuous gap: sampling a synthetic MDP is as feasible as sampling a synthetic tabular
dataset, yet no in-context RL work treats prior design as a primary objective. Second, MDPs admit a fixed-size sufficient statistic, independent of the episodes observed and tabular in shape, which makes them directly amenable to the
attention-based architectures used for tabular foundation models, with a policy head replacing the supervised target. Together these define the agenda for an RL foundation model. As a proof of concept, we train a Graph
Attention Network entirely on synthetic MDPs and show that, with no task-specific tuning, it solves held-out tabular
benchmarks in context, both online and offline: online, in far fewer episodes than UCB-VI and tabular Q-learning, and offline, competitively with VI-LCB.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.IR
Rescaling MLM-Head for Neural Sparse Retrieval
Youngjoon Jang, Seongtae Hong, Jonah Turner et al. (+1 more)
Learned sparse
retrieval (LSR) models such as SPLADE have traditionally used BERT-style masked
language models as backbone encoders. A natural expectation is that replacing BERT with stronger pretrained encoders should improve
retrieval effectiveness. However, we find that under standard SPLADE training recipes, backbones with large MLM-head L2 norms can suffer performance degradation and even training collapse under standard SPLADE training recipes. We identify this failure as a scale mismatch in the MLM head: SPLADE directly uses MLM-head outputs to construct sparse lexical representations, and query-document relevance is computed by an unnormalized dot product over these representations. As a result, an inflated MLM-head scale can amplify sparse activations, distort matching scores, and destabilize contrastive training under common training settings. To address this issue, we introduce a simple initialization-time correction that rescales the MLM-head projection by a constant factor before SPLADE training. This zero-cost adjustment improves training stability without modifying the model architecture or training objective. Across both in-domain and out-of-domain
retrieval benchmarks, this simple correction substantially improves large-norm backbones such as ModernBERT and Ettin, turning unstable training runs into competitive sparse retrievers. In several settings, the corrected models further match or surpass the classic BERT-SPLADE baseline. These findings suggest that the bottleneck in adapting pretrained encoders to LSR is not encoder capacity alone, but the calibration of the MLM-head scale used to construct sparse lexical representations.
Read the full abstract ▾
All authors Youngjoon Jang, Seongtae Hong, Jonah Turner, Heuiseok Lim
Artificial Intelligence · Cross-listed from cs.LG
Learning from Own Solutions: Self-Conditioned Credit Assignment for Reinforcement Learning with Verifiable Rewards
Yingyu Shan, Yuhang Guo, Zihao Cheng et al. (+7 more)
Reinforcement learning with verifiable rewards (RLVR) has driven substantial progress in training LLMs for
reasoning tasks, but representative methods such as GRPO assign uniform credit across all
tokens, wasting gradient on routine
tokens while under-crediting pivotal
reasoning steps. Existing
token-level credit assignment methods require resources beyond the model's own rollouts. GRPO variants rely on process reward models or ground-truth answers. Knowledge
distillation assigns credit through per-
token divergence but requires external teachers (On-Policy
Distillation) or privileged information (On-Policy Self
Distillation). However, these dependencies limit applicability in the pure RLVR setting. We observe that conditioning the model on its own verified trajectories induces a measurable per-
token KL divergence between the original and conditioned distributions, and prove that distilling from a self-teacher constructed by verified trajectories leads to infeasible weighted-average solutions when multiple verified trajectories exist. We propose SC-GRPO (Self-Conditioned GRPO), which uses KL divergence mentioned before as a multiplicative weight on GRPO gradients. Across five
benchmarks spanning math, code, and agentic tasks, SC-GRPO consistently outperforms 8.1% over GRPO and 5.9% over DAPO with stronger OOD performance. Moreover, SC-GRPO achieves higher performance than OPD.
Read the full abstract ▾
All authors Yingyu Shan, Yuhang Guo, Zihao Cheng, Zeming Liu, Xiangrong Zhu, Xinyi Wang, Jiashu Yao, Wei Lin, Hongru Wang, Heyan Huang
Artificial Intelligence
ProfiLLM: Utility-Aligned Agentic User Profiling for Industrial Ride-Hailing Dispatch
Tengfei Lyu, Zirui Yuan, Xu Liu et al. (+4 more)
Bringing Large
Language Models (LLMs) into industrial ride-hailing dispatch as semantic feature extractors over platform-scale behavioral logs is a compelling but under-explored data systems problem. Production matching pipelines remain dominated by structured numerical features, yet decisive behavioral signals (e.g., a driver's habitual aversion to certain regions) are inherently contextual and naturally expressible as LLM-generated user profiles. However, scaling such profiling to a live, millisecond-latency dispatcher faces three intertwined constraints rarely addressed together: on a platform with millions of daily orders, logs exceed any LLM's context window by orders of magnitude; most users are long-tail, with too few interactions for per-user profiling; and surface-fluent profiles do not necessarily improve downstream prediction utility. We present ProfiLLM, an agentic LLM data pipeline that operationalizes utility-aligned user profiling for production matching systems through two modules. (1) Tool-Augmented Global Knowledge Mining equips an LLM
agent with 27 analytical tools to mine platform-scale data, producing reusable global knowledge, adaptive user clustering rules, and region-level supply-demand priors. (2) Utility-Aligned Profile Exploration generates multiple candidate profiles per cluster, evaluates them via a lightweight downstream utility proxy, iteratively refines the best candidates and constructs preference pairs for DPO fine-tuning. Deployed on DiDi's production dispatcher, ProfiLLM achieves up to +6.14% relative AUC improvement in outcome prediction, up to +4.35% GMV gain in dispatching simulation, and consistent improvements in a 14-day online A/B test including +0.47% GMV, +0.33% Completion Rate, and -0.82% Cancel-Before-Accept rate.
Read the full abstract ▾
All authors Tengfei Lyu, Zirui Yuan, Xu Liu, Kai Wan, Zihao Lu, Li Ma, Hao Liu
Artificial Intelligence · Cross-listed from cs.IR
SHIFT: Semantic Harmonization via Index-side Feature Transformation for Multilingual Information Retrieval
Youngjoon Jang, Seongtae Hong, Hyeonseok Moon et al. (+1 more)
With the rapid expansion of massive
multilingual corpora,
Multilingual Information
Retrieval (MLIR) has emerged as a critical technology for global information access. MLIR enables users to retrieve semantically relevant documents from
multilingual text collections using a single-
language query. However, recent
multilingual dense
retrieval models often exhibit a strong preference for documents in the same
language as the query. This leads to severe
language bias, where top-ranked results are dominated by documents of specific
languages, even when documents in other
languages contain more semantically relevant information. To address this issue, we propose SHIFT, a training-free method applicable in the indexing stage. Specifically, SHIFT utilizes parallel translation pairs to estimate a relative
language vector for each target
language with respect to a source
language. Subsequently, SHIFT corrects the
language-specific offset by subtracting this relative
language vector from document
embeddings during indexing. Our comprehensive evaluation across four MLIR
benchmarks and diverse dense
retrieval models confirms that SHIFT can effectively mitigate
language bias and enhance MLIR performance.
Read the full abstract ▾
All authors Youngjoon Jang, Seongtae Hong, Hyeonseok Moon, Heuiseok Lim
Artificial Intelligence · Cross-listed from cs.SD
Closing the Loop: PID Feedback Control for Interpretable Activation Steering in Symbolic Music Generation
Ioannis Prokopiou, Pantelis Vikatos, Maximos Kaliakatsos-Papakostas et al. (+2 more)
Transformer-based architectures have significantly advanced the generation of complex symbolic sequences, yet a significant gap remains in achieving fine-grained, interpretable control over discrete signal attributes. This paper investigates the mechanistic interpretability of the Multitrack Music
Transformer (MMT) and proposes a framework for deterministic attribute modulation without retraining to bridge this gap via inference-time activation steering. Utilizing the Difference-in-Means (DiffMean) methodology, we isolate latent directions for signal attributes, specifically Pitch and Duration, within the residual stream. We validate the Linear Representation Hypothesis in this domain, achieving high correlation between steering magnitude and attribute shift. To address the inherent feature
entanglement in multi-attribute steering, we introduce a Dual Steering framework utilizing Gram-Schmidt Orthogonalization. Experimental results demonstrate that this
geometric decoupling reduces conceptual interference and signal degradation compared to naive vector addition, enabling independent deterministic control even against strong autoregressive conditioning.
Read the full abstract ▾
All authors Ioannis Prokopiou, Pantelis Vikatos, Maximos Kaliakatsos-Papakostas, Theodoros Giannakopoulos, Themos Stafylakis
Artificial Intelligence
R2D-RL: A RoboCup 2D Soccer Environment for Multi-Agent Reinforcement Learning
Haobin Qin, Baofeng Zhang, Hidehisa Akiyama et al. (+1 more)
Robot soccer is a challenging testbed for multi-
agent reinforcement learning because it combines partial observability, cooperative and
adversarial interaction, sparse rewards, and long-horizon tactical behavior. RoboCup 2D Soccer Simulation (RCSS2D) provides a mature
robot-soccer platform, but its competition-oriented server-client architecture is difficult to use directly with modern Python-based MARL workflows. We introduce R2D-RL, a reinforcement learning environment that connects RCSS2D and HELIOS-based player clients to a Python MARL interface through shared-
memory communication and cycle-level synchronization. R2D-RL supports full-field and scenario-based training with configurable opponents, Base discrete and Hybrid parameterized action spaces, action masks, expected possession value (EPV)-based reward shaping, and parallel execution. We provide front-goal scenarios and an 11-vs-11 full-field
benchmark, together with baseline results.
Read the full abstract ▾
All authors Haobin Qin, Baofeng Zhang, Hidehisa Akiyama, Keisuke Fujii
Artificial Intelligence · Cross-listed from cs.LG
Bayesian Anytime Pareto Set Identification for Multi-Objective Multi-Armed Bandits
Lennert Saerens, Bram Silue, Eleni Litsa et al. (+2 more)
Identifying Pareto optimal solutions is critical to support multi-objective decision-making. In this work, we present Top-Two Pareto Front Thompson Sampling (TTPFTS), the first anytime
algorithm for Pareto Set Identification in Multi-Objective Multi-Armed Bandits (MOMABs), taking a
Bayesian approach. We
benchmark TTPFTS against state-of-the-art fixed-confidence and fixed-budget Pareto Set Identification
algorithms on synthetic environments. Next, we demonstrate its practical utility in a challenging real-world multi-objective multi-
agent molecular discovery setting by efficiently exploring an ultra-large synthesis-on-demand molecular library, outperforming industry-standard and state-of-the-art baselines. Furthermore, we introduce a novel uncertainty quantification metric that estimates our
algorithm's confidence in the predicted Pareto set. We demonstrate that this metric effectively proxies true performance, yielding a robust methodology for monitoring learning progress in complex settings. Finally, we complement these empirical findings with the first finite-time theoretical analysis of an anytime MOMAB Pareto Set Identification
algorithm.
Read the full abstract ▾
All authors Lennert Saerens, Bram Silue, Eleni Litsa, Peter Vrancx, Pieter Libin
Artificial Intelligence · Cross-listed from cs.CL
RedactionBench
Sean Brynjólfsson, Shashvat Jayakrishnan, Esha Sali et al. (+2 more)
Large
Language Models are increasingly applied to sensitive domains that require redaction of personally identifiable information (PII). While redacting PII is a data cleaning prerequisite, existing
benchmarks conflate extraction mechanics with
privacy semantics. A public phone number is not equivalent to a phone number in a medical record. Whether information constitutes a violation depends heavily on who holds it, why, and in what context, fundamentally differentiating redaction from simple entity recognition. Grounded in contextual integrity, we introduce RedactionBench, a manually annotated
benchmark comprising 200 diverse documents across 11 domains, mostly seeded from real-world sources. We also introduce R-Score, a novel character-level metric that treats semantically similar redactions equally and nullifies shallow formatting choices, such as varying masking styles for phone numbers. Evaluations across Named Entity Recognition models, entity extraction Small
Language Models, and frontier models equipped with agentic tools demonstrate that contextual redaction remains an unsolved problem. A human evaluation with over 80 users on RedactionBench reveals a stark dichotomy in
privacy perceptions. Annotators show consensus with target labels for mandatory redactions (89.4 percent) and safe text preservations (94.1 percent), but fail to agree on contextual redactions (47.7 percent). This variance demonstrates the subjective nature of contextual
privacy and motivates R-Score, which decouples contextual ambiguity from strict precision. We compare 35 models across families and report their performance in redacting PII. Finally, we release RedactionBench to establish a baseline for future
privacy-preserving systems, hoping to inspire efficient model design and standardized evaluations.
Read the full abstract ▾
All authors Sean Brynjólfsson, Shashvat Jayakrishnan, Esha Sali, Diptanshu Purwar, Madhav Aggarwal