Temporal Video Grounding (TVG) aims to precisely localize video segments corresponding to natural language queries, which is a critical capability for long-form video understanding. Although existing reinforcement learning approaches encourage models to generate reasoning chains before predictions, they fail to explicitly constrain the reasoning process to ensure the quality of the final temporal predictions. To address this limitation, we propose Timestamp Anchor-constrained Reasoning for Temporal Video Grounding (TAR-TVG), a novel framework that introduces timestamp anchors within the reasoning process to enforce explicit supervision to the thought content. These anchors serve as intermediate verification points. More importantly, we require each reasoning step to produce increasingly accurate temporal estimations, thereby ensuring that the reasoning process contributes meaningfully to the final prediction. To address the challenge of low-probability anchor generation in models (e.g., Qwen2.5-VL-3B), we develop an efficient self-distillation training strategy: (1) initial GRPO training to collect 30K high-quality reasoning traces containing multiple timestamp anchors, (2) supervised fine-tuning (SFT) on distilled data, and (3) final GRPO optimization on the SFT-enhanced model. This three-stage training strategy enables robust anchor generation while maintaining reasoning quality. Experiments show that our model achieves state-of-the-art performance while producing interpretable, verifiable reasoning chains with progressively refined temporal estimations.

Addressing intricate real-world problems necessitates in-depth information seeking and multi-step reasoning. Recent progress in agentic systems, exemplified by Deep Research, underscores the potential for autonomous multi-step research. In this work, we present a cohesive paradigm for building end-to-end agentic information seeking agents from a data-centric and training-stage perspective. Our approach consists of four key stages: (1) browsing data construction, (2) trajectories sampling, (3) supervised fine-tuning for effective cold start, and (4) reinforcement learning for enhanced generalisation. We instantiate this framework in a web agent based on the ReAct, WebDancer. Empirical evaluations on the challenging information seeking benchmarks, GAIA and WebWalkerQA, demonstrate the strong performance of WebDancer, achieving considerable results and highlighting the efficacy of our training paradigm. Further analysis of agent training provides valuable insights and actionable, systematic pathways for developing more capable agentic models. The codes and demo will be released in https://github.com/Alibaba-NLP/WebAgent.

In order to work more efficiently, accurately, reliably, and safely in industrial scenarios, robots should have at least general knowledge, working-environment knowledge, and operating-object knowledge. These pose significant challenges to existing embodied intelligent robotics (EIR) techniques. Thus, this paper first briefly reviews the history of industrial robotics and analyzes the limitations of mainstream EIR frameworks. Then, a knowledge-driven technical framework of embodied intelligent industrial robotics (EIIR) is proposed for various industrial environments. It has five modules: a world model, a high-level task planner, a low-level skill controller, a simulator, and a physical system. The development of techniques related to each module are also thoroughly reviewed, and recent progress regarding their adaption to industrial applications are discussed. A case study is given to demonstrate the newly proposed EIIR framework's applicability to real-world assembly system. Finally, the key challenges that EIIR encounters in industrial scenarios are summarized and future research directions are suggested. The authors believe that EIIR technology is shaping the next generation of industrial robotics and EIIR-based industrial systems supply a new technological paradigm for intelligent manufacturing. It is expected that this review could serve as a valuable reference for scholars and engineers that are interested in industrial embodied intelligence. Together, scholars can use this research to drive their rapid advancement and application of EIIR techniques. The interested authors would continue to track and contribute new studies in the project page https://github.com/jackyzengl/EIIR.

How do we predict others from patterns in their behavior and what are the computational constraints that limit this ability? We investigate these questions by modeling human behavior over repeated games of rock, paper, scissors from Brockbank & Vul (2024). Against algorithmic opponents that varied in strategic sophistication, people readily exploit simple transition patterns (e.g., consistently playing rock after paper) but struggle to detect more complex sequential dependencies. To understand the cognitive mechanisms underlying these abilities and their limitations, we deploy Hypothetical Minds (HM), a large language model-based agent that generates and tests hypotheses about opponent strategies, as a cognitive model of this behavior (Cross et al., 2024). We show that when applied to the same experimental conditions, HM closely mirrors human performance patterns, succeeding and failing in similar ways. To better understand the source of HM's failures and whether people might face similar cognitive bottlenecks in this context, we performed a series of ablations and augmentations targeting different components of the system. When provided with natural language descriptions of the opponents' strategies, HM successfully exploited 6/7 bot opponents with win rates >80% suggesting that accurate hypothesis generation is the primary cognitive bottleneck in this task. Further, by systematically manipulating the model's hypotheses through pedagogically-inspired interventions, we find that the model substantially updates its causal understanding of opponent behavior, revealing how model-based analyses can produce testable hypotheses about human cognition.

The inference-time resource costs of large language and vision models present a growing challenge in production deployments. We propose the use of foundation model programs, i.e., programs that can invoke foundation models with varying resource costs and performance, as an approach to this problem. Specifically, we present a method that translates a task into a program, then learns a policy for resource allocation that, on each input, selects foundation model "backends" for each program module. The policy uses smaller, cheaper backends to handle simpler subtasks, while allowing more complex subtasks to leverage larger, more capable models. We evaluate the method on two new "streaming" visual question-answering tasks in which a system answers a question on a sequence of inputs, receiving ground-truth feedback after each answer. Compared to monolithic multi-modal models, our implementation achieves up to 98% resource savings with minimal accuracy loss, demonstrating its potential for scalable and resource-efficient multi-modal inference.

Autonomous vehicle navigation in structured environments requires planners capable of generating time-optimal, collision-free trajectories that satisfy dynamic and kinematic constraints. We introduce V*, a graph-based motion planner that represents speed and direction as explicit state variables within a discretised space-time-velocity lattice. Unlike traditional methods that decouple spatial search from dynamic feasibility or rely on post-hoc smoothing, V* integrates both motion dimensions directly into graph construction through dynamic graph generation during search expansion. T o manage the complexity of high-dimensional search, we employ a hexagonal discretisation strategy and provide formal mathematical proofs establishing optimal waypoint spacing and minimal node redundancy under constrained heading transitions for velocity-aware motion planning. We develop a mathematical formulation for transient steering dynamics in the kinematic bicycle model, modelling steering angle convergence with exponential behaviour, and deriving the relationship for convergence rate parameters. We further demonstrate V*'s performance in simulation studies with cluttered and dynamic environments involving moving obstacles, showing its ability to avoid conflicts, yield proactively, and generate safe, efficient trajectories with temporal reasoning capabilities for waiting behaviours and dynamic coordination. Autonomous navigation requires planning algorithms that can compute time-efficient and dynamically feasible trajectories. Among the diverse approaches to motion planning, optimal pathfinding algorithms are recognized for their ability to guarantee high-quality solutions by minimizing path costs under strict constraints. This makes them particularly valuable in environments where precision and efficiency are critical. Classical pathfinding methods such as the Dijkstra ( 1) and A* ( 2) algorithms have been widely adopted for solving shortest path problems on graphs due to their efficiency.

Geometry problem solving (GPS) requires models to master diagram comprehension, logical reasoning, knowledge application, numerical computation, and auxiliary line construction. This presents a significant challenge for Multimodal Large Language Models (MLLMs). However, existing benchmarks for evaluating MLLM geometry skills overlook auxiliary line construction and lack fine-grained process evaluation, making them insufficient for assessing MLLMs' long-step reasoning abilities. To bridge these gaps, we present the GeoLaux benchmark, comprising 2,186 geometry problems, incorporating both calculation and proving questions. Notably, the problems require an average of 6.51 reasoning steps, with a maximum of 24 steps, and 41.8% of them need auxiliary line construction. Building on the dataset, we design a novel five-dimensional evaluation strategy assessing answer correctness, process correctness, process quality, auxiliary line impact, and error causes. Extensive experiments on 13 leading MLLMs (including thinking models and non-thinking models) yield three pivotal findings: First, models exhibit substantial performance degradation in extended reasoning steps (nine models demonstrate over 50% performance drop). Second, compared to calculation problems, MLLMs tend to take shortcuts when solving proving problems. Third, models lack auxiliary line awareness, and enhancing this capability proves particularly beneficial for overall geometry reasoning improvement. These findings establish GeoLaux as both a benchmark for evaluating MLLMs' long-step geometric reasoning with auxiliary lines and a guide for capability advancement. Our dataset and code are included in supplementary materials and will be released.

Large reasoning models (LRMs) achieve impressive reasoning capabilities by generating lengthy chain-of-thoughts, but this "over-thinking" incurs high latency and cost without commensurate accuracy gains. In this work, we introduce AALC, a lightweight, accuracy-aware length reward integrated into reinforcement learning that dynamically balances correctness and brevity during training. By incorporating validation accuracy into the reward and employing a smooth, dynamically scheduled length penalty, AALC delays length penalty until target performance is met. Through extensive experiments across standard and out-of-distribution math benchmarks, we show that our approach reduces response length by over 50% while maintaining or even improving the original accuracy. Furthermore, qualitative analysis reveals that our method curbs redundant reasoning patterns such as excessive subgoal setting and verification, leading to structurally refined outputs rather than naive truncation. We also identify that efficiency gains are accompanied by reduced interpretability: models trained with AALC omit some narrative framing and explanatory context. These findings highlight the potential of reward-based strategies to guide LRMs toward more efficient, generalizable reasoning paths.

Temporal Table Reasoning is a critical challenge for Large Language Models (LLMs), requiring effective reasoning to extract relevant insights. Despite existence of multiple prompting methods, their impact on table reasoning remains largely unexplored. Furthermore, model performance varies drastically across different table and context structures, making it difficult to determine an optimal approach. This work investigates multiple prompting technique on diverse table types to determine that performance depends on factors such as entity type, table structure, requirement of additional context and question complexity, with "NO" single method consistently outperforming others. To address this, we introduce SEAR, an adaptive prompting framework inspired by human reasoning that dynamically adjusts to context and integrates structured reasoning. Our results demonstrate that SEAR achieves superior performance across all table types compared to baseline prompting techniques. Additionally, we explore the impact of table structure refactoring, finding that a unified representation enhances model reasoning.

Reasoning with minimal models has always been at the core of many knowledge representation techniques, but we still have only a limited understanding of this problem in Description Logics (DLs). Minimization of some selected predicates, letting the remaining predicates vary or be fixed, as proposed in circumscription, has been explored and exhibits high complexity. The case of `pure' minimal models, where the extension of all predicates must be minimal, has remained largely uncharted. We address this problem in popular DLs and obtain surprisingly negative results: concept satisfiability in minimal models is undecidable already for $\mathcal{EL}$. This undecidability also extends to a very restricted fragment of tuple-generating dependencies. To regain decidability, we impose acyclicity conditions on the TBox that bring the worst-case complexity below double exponential time and allow us to establish a connection with the recently studied pointwise circumscription; we also derive results in data complexity. We conclude with a brief excursion to the DL-Lite family, where a positive result was known for DL-Lite$_{\text{core}}$, but our investigation establishes ExpSpace-hardness already for its extension DL-Lite$_{\text{horn}}$.