Search arXiv⌕ Search

arXiv subjects

Ji-Rong Wen

Publications and source records attributed to Ji-Rong Wen.

At least 19 recordsLinked to original sources
arXiv2026

ToolSpace: Scaling Tool-Centric Environment Synthesis for Long-Horizon Agents

Large language models are deployed as long-horizon agents that compose diverse tools across MCP services into executable workflows. Existing MCP environment-scaling methods are data-centric: they first synthesize a database and scenario, then derive tools and tasks from them. As a result, tool interfaces may not match real ones, tool compositions are not grounded in execution, and tasks rarely span more than one environment. We propose ToolSpace, a tool-centric framework that reverses this order. ToolSpace mines data, state, and retry dependencies from adjacent calls in execution traces of real MCP services and open agent corpora, using deterministic rules rather than LLM judgment, and clusters the resulting weighted tool-composition graph into environments. Databases are synthesized only for state-mutating environments, while query and stateless tools replay recorded returns. ToolSpace then performs state-aware random walks with rollback and decompiles each chain into a natural-language task with assertion-based checks, keeping only samples that replay successfully from the initial state. Applied to AutomationBench and tau3-Bench Banking, the pipeline produces harder Pro versions on which 7 frontier models drop by up to 38.2 points. Across seven long-horizon tool-use benchmarks, ToolSpace raises Qwen3.6-35B-A3B from 23.0 to 41.0 average score (43.2 with RL) and GLM-5.2 from 44.6 to 56.6, on par with Claude Opus 5. Ablations show that execution-witnessed edges clearly outperform LLM-judged ones, and performance keeps improving with data and environment scale.

cs.CL↗
arXiv2026

CoworkOS:Scaling Agent Coworkers for Real-world Automation

Rapid advances in cowork agents are shifting human-AI collaboration toward real-world automation, delivering software, designs, and other artifacts. However, such productivity work demands both high-quality delivery on complex productivity tasks and continued monitoring and decision-making over long horizons, a combination that remains an open challenge. We introduce CoworkOS, a general-purpose harness for long-horizon work that separates task strategy from execution management: the model decides actions, delegation, and acceptance, while the runtime maintains state and feedback. Its three-layer architecture enables direct execution or delegation to Subagents, synchronously or asynchronously, under stable responsibilities. Through a shared application layer, humans and agents inspect and revise the same evolving documents, presentations, CAD models, and Blender scenes, so human-in-the-loop participation can enter both through conversation and through direct modification of the work itself. To sustain continuity across agents and sessions, CoworkOS further introduces hierarchical workspaces and persistent semantic memory, so working files, key decisions, and long-running Job logs remain shareable, reusable, and resumable after interruption. Across 12 benchmarks and 6 categories of real-world applications, CoworkOS proves highly competitive with frontier agent applications such as Codex and DeepSeek Harness. As a sustained-work case study, CoworkOS uses GLM-5.3 as its backbone to conduct RSI-driven automated post-training on Qwen3.6-35B-A3B. Over five rounds, the model's average score across eight benchmarks rose from 33.19 to 42.95 (+9.76). Looking ahead, we believe CoworkOS opens a meaningful path toward general cowork-agent intelligence.

cs.AI↗
arXiv2026

CalibForge: Adversarial Solver Calibration for Scaling Learnable Terminal Tasks

Training terminal agents requires executable and verifiable tasks that are not merely solvable, but appropriately challenging for learning. Executable validation establishes feasibility, yet does not reveal how a task behaves relative to a given solver setting. In this paper, we present CalibForge, an autonomous terminal-task synthesis system that uses verified solver behavior to revise candidate tasks through adversarial solver calibration. Multi-solver calibration targets disagreement within a heterogeneous solver pool, whereas contrastive solver calibration targets a designated strong-pass/weak-fail relation; both operationalize a solver-relative learnable zone anchored in demonstrated solvability. Using CalibForge, we construct 5,431 calibrated terminal tasks. Our ablations show that both strategies yield more effective supervision than authoring and validation alone or ordinary single-solver feedback. Models trained on the full collection achieve 31.84% and 47.57% on Terminal-Bench 2.1. The largest improvements over the corresponding base model reach 23.97 percentage points on Terminal-Bench 2.1, 26.54 points on SWE-bench Pro, and 30.04 points on Doc2Repo. Together, these results support solver-relative learnability as a practical target for constructing effective and transferable agent training data.

cs.LG↗
arXiv2026

Harness Scaling for Generalizable Agentic Reinforcement Learning

Agent harnesses have become a central component of modern agentic systems, shaping how models use tools, manage context, and coordinate long-horizon interaction. Realizing the benefits of these harnesses depends on how effectively the underlying model can make use of them, motivating reinforcement learning directly through the harness. While recent efforts have explored harness diversification to reduce specialization to particular execution settings, how to systematically construct and organize harness diversity for cross-harness generalization remains insufficiently understood. This work studies Harness Scaling as a distinct scaling dimension for agentic RL, focusing on structured variation in how policies execute and interact with the same underlying task and environment. Concretely, we construct a factorized harness space, select a compact training subset with balanced factorial coverage, and organize cross-harness training through locality-aware task assignment. Across multiple benchmarks, model scales, and RL algorithms, Harness Scaling consistently improves generalization to unseen harness combinations. Moreover, policies trained through structured white-box harness variation transfer effectively to unseen external black-box harnesses, outperforming policies trained through a single complex black-box harness. These results establish harness diversity as a meaningful scaling dimension for agentic RL and highlight the importance of how this diversity is organized for cross-harness generalization.

cs.LG↗
arXiv2026

Evolving with Open Worlds: Towards Recursive Self-Improvement for Future Forecasting

Most contemporary artificial intelligence (AI) systems are released with frozen parameters and improve through external updates, whereas an ideal system would keep improving from its own experience, a capacity we call recursive self-improvement (RSI). Yet studying RSI requires an environment that is continuously changing, objectively verifiable, and naturally grounded, posing a fundamental challenge for agent evolution. Here we introduce EvoForesight, a sustainable evolution agent system grounded in future event prediction, in which agents continuously interact with a dynamically evolving environment and improve by internalizing and externalizing the experience gained through these interactions. A living environment synthesizes verifiable tasks, and a recursive-evolution engine turns the reasoning trajectory behind each prediction back into capability: decoupling search from prediction adapts parameter-level evolution to the contamination-prone live environment, while distilling reusable skills lets the agent evolve from the limited live trajectories such an environment affords. On the external FutureX leaderboard, which evaluates predictions against real-world outcomes and is updated weekly, our self-improved forecaster ranked first across all systems over three consecutive weeks, achieving up to a 195% improvement in rating. On our synthesized evaluation, it ranked among the top three across four base LLMs, outperforming the strongest baseline by up to 10.7 Bradley--Terry points. More broadly, these results suggest that a system built for self-improvement is best understood as an agent that co-evolves with a genuinely living, open-ended environment, pointing to a shift in AI from training static models on fixed data toward cultivating agents that grow stronger from their own experience, with such an open world as the indispensable basis of that growth.

cs.CL↗
arXiv2026

OpenSaC: A Python Library for Search as Code in Deep Research Agents

Deep research agents investigate complex questions through iterative search, source examination, and evidence synthesis. A conventional workflow alternates reasoning with individual search and reading actions. Search as Code introduces a new paradigm: agents write programs that compose search, source reading, reranking, model calls, and local processing, returning selected observations for further reasoning. To support this workflow in existing coding agents, we present OpenSaC, an open-source Python library with a companion Search as Code skill. Configurable capability interfaces let agents combine these operations while retaining evidence for inspection and reuse. Compared with standard Codex using its built-in web search, the OpenSaC configuration uses less execution time and fewer model tokens on BrowseComp and WideSearch, with higher accuracy on BrowseComp and slightly lower row F1 on WideSearch. We aim to lower the barrier to adopting Search as Code and provide an open foundation for community development and transparent research on deep research agents.

cs.AI↗
arXiv2026

Agent-Reflex: Teaching Language Agents to Act Reflectively

Large language models have evolved from single-turn chatbots into interactive agents capable of long-horizon reasoning. In real deployments, MCP tool services can return noisy feedback and users may intervene to correct the agent, but existing methods provide limited supervision for reflection learning and failure recovery. To bridge this gap, we propose Agent-Reflex, a general training framework for incentivizing proactive reflection in agents. Specifically, Agent-Reflex comprises three stages. First, Environment Reflection Skill Discovery (RSD) injects grounded tool-feedback perturbations into environment scaling, samples a capability-diverse LLM pool, mines matched success-failure pairs, and summarizes reusable reflection meta-skills. Building on this skill library, we then apply Reflective Supervised Fine-Tuning (ReSFT) that converts each skill into supervision for self-reflection, skill recovery, and user-centric correction, strengthening complementary reflective behaviors. Finally, we propose On-Policy Reflective Policy Optimization (ORPO) to combine agentic reinforcement learning under noisy interactions with reflection-guided on-policy distillation to further elicit proactive reflection. Built on Qwen3-8B, Agent-Reflex reaches 58.6 overall on five long-horizon tool-use benchmarks and 29.1 on AgentNoiseBench, surpassing strong environment-scaling and reflection baselines. Experiments on six challenging agent benchmark suites demonstrate the effectiveness of Agent-Reflex. Further analyses show that Agent-Reflex improves agents' robustness to noise while maintaining stable training dynamics, supporting scalable agent training. Our code is available at https://github.com/RUC-NLPIR/Agent-Reflex.

cs.CL↗
arXiv2026

FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents

Training deep-search agents requires verifiable questions that demand substantial evidence acquisition, yet structurally complex tasks may still admit cheap identifying routes when clues are overly selective, co-covered by the same evidence, directly searchable, or resolved from solver priors. We formalize this gap with a shortcut-aware difficulty framework that distinguishes intended construction complexity from identifying routes available through retrieval. Guided by this framework, we introduce FORT, a Framework of Shortcut-Resistant Training-Data Synthesis, which targets four shortcut risks across entity selection, evidence graph construction, question formulation, and adversarial refinement. We evaluate realized search behavior with solver-conditioned trajectory diagnostics rather than graph complexity or search length alone. Compared with existing open-source deep-search datasets, FORT shows later answer exposure, higher retrieval effort, lower single-clue selectivity, greater evidence dispersion, and higher realized pre-answer dependency costs. Using the resulting trajectories, we train FORT-Searcher with supervised fine-tuning alone. Across model scales, FORT-Searcher consistently improves over its backbones, achieving the strongest overall performance among similarly sized agents. Relevant resources will be made available at https://github.com/RUCAIBox/FORT-Searcher.

cs.CL↗
arXiv2026

Spectral-Driven Adaptive-Rank Optimizer for Large Language Model Reinforcement Learning with Verifiable Rewards

Reinforcement learning with verifiable rewards (RLVR) is reshaping how large language models (LLMs) acquire reasoning capabilities, yet the optimizers used remain largely inherited from teacher-forcing training. This raises a fundamental question: does RLVR induce a distinct optimization geometry that calls for a different optimizer? We investigate this question through the spectral structure of LLM gradients and find that gradient energy progressively concentrates into a shrinking effective subspace, which is characteristic of RLVR, as it is absent under supervised fine-tuning (SFT) and is largely removed when the verifiable reward is replaced by a random one. Moreover, the rank required to capture a fixed fraction of gradient energy steadily decreases over training, revealing that the effective dimensionality of useful updates itself evolves with training. These findings motivate a new optimization principle: rather than prescribing the update rank externally, an optimizer should adapt its rank to the evolving spectral structure of the gradient. We introduce Ares (Adaptive Rank-Effective Spectral Optimization), which estimates the effective update rank from the gradient's spectral energy distribution, tracks its evolution, and confines the gradient signal driving each update to the resulting low-rank subspace. Across LLMs of different scales, Ares improves accuracy over widely used optimizers and generalizes across RLVR algorithms and tasks. Results suggest that effective optimization may require optimizers to be designed around the geometry and dynamics of the training process. We release the code and data at https://github.com/RUCAIBox/Ares.

cs.CL↗
arXiv2026

Gestalt: Large Multimodal Interplay Model

In this paper, we propose Gestalt, a new paradigm of large multimodal model built around multimodal interplay. Despite rapid advances, large multimodal models are reaching a bottleneck: existing approaches focus primarily on accommodating additional modalities while overlooking the distinct characteristics of each modality and the relations among them. Motivated by the multistage property of human multisensory perception, we propose a multimodal interplay pyramid that organizes multimodal modeling as a progression from modality-specific processing, through cross-modal alignment, to deeper multimodal integration. Guided by this pyramid, Gestalt adopts a unified discrete diffusion framework and an interplay-partitioned architecture, with learnable interplay tokens mediating cross-modal exchange and integration. The pyramid also structures its data organization and training strategy. Strong performance across image generation, multimodal understanding, and text-only evaluation shows that Gestalt significantly improves cross-modal integration while preserving modality-specific information, effectively harnessing the strengths of diffusion-based multimodal models and offering a promising path toward unified multimodal intelligence. Project page: https://GeWu-Lab.github.io/Gestalt.

cs.CV↗
arXiv2026

When Should Agents Think? Adaptive Reasoning via Cross-Turn Estimation

Large language model (LLM)-based agents have demonstrated strong capabilities on complex tasks. They typically perform reasoning before each action throughout an interaction trajectory. However, reasoning may not be necessary at every turn, as reasoning produced earlier can continue to support subsequent actions. A key challenge is therefore to determine when existing reasoning remains sufficient and when a new reasoning step is needed, without relying on costly generation-based verification. We find that decreases in the likelihood of subsequent reference actions after removing additional reasoning closely track whether those actions remain recoverable given earlier reasoning, providing an effective and lightweight signal for estimating cross-turn action support. Based on this observation, we propose Reasoning Adaptation through Cross-Turn Estimation (RACE), a training approach for adaptive agent reasoning. RACE introduces a Likelihood-Guided Progressive Reasoning Cover Detection (LoGiC) procedure that progressively identifies reasoning turns whose removal has limited impact on the current and subsequent reference actions. The resulting removal signals are incorporated into both supervised fine-tuning and agentic reinforcement learning, enabling the policy to learn when to reason and when to act directly. Extensive experiments on four representative agent benchmarks show that RACE substantially reduces reasoning cost while maintaining or improving task performance.

cs.AI↗
arXiv2026

PhoneWorld: From Real-App Trajectories to Dynamic and Verifiable Environments for Phone-Use Agents

Real applications provide the training setting closest to phone-agent deployment, but are difficult to reset, scale safely, and verify programmatically. Static screenshots and interaction trajectories preserve realistic evidence but cannot generate new experience. We introduce PhoneWorld, a trace-grounded framework that converts such evidence into runnable, resettable, and verifiable Android environments. PhoneWorld induces a usage-weighted interaction skeleton from observed pages, transitions, and state-changing operations; translates it into a behavior-grounded app specification; realizes the specification through an autonomous build--inspect--repair loop; and synthesizes executable tasks with programmatic verifiers. The resulting suite spans 34 consumer-facing apps across 16 domains and supports an audited online benchmark, verified trajectory generation, and online RL through common reset and verification interfaces. Evaluations with diverse general and open-source GUI agents show that PhoneWorld supports reliable end-to-end online interaction and exposes capabilities complementary to AndroidWorld. Controlled SFT experiments further show that PhoneWorld trajectories complement AndroidWorld supervision, transfer across online and offline benchmarks, and become more effective as data volume and app coverage increase. Under a matched RL budget, combining PhoneWorld mock-app rollouts with real-app rollouts improves performance over real-app RL alone on both real-phone tasks and AndroidWorld. Together, these results demonstrate that trace-grounded executable abstraction can bridge realistic mobile behavior and scalable agent learning, turning limited real-app evidence into a growing supply of controllable and verifiable environments for training and evaluation.

cs.CL↗
arXiv2026

Agent-Editing World Model: Rethinking World Modeling for LLM Agents

Recent advances in large language models (LLMs) have enabled agents to tackle long-horizon tasks across diverse environments. To further improve agent performance, existing language world models typically predict environment observations, yet reconstructing high-entropy, execution-dependent tool responses offers limited value when real feedback is available. Meanwhile, agents suffer from \emph{task-state contamination}, where unsupported assumptions and outdated plans persist in history and distort subsequent decisions. We propose the \textbf{Agent-Editing World Model (AEWM)}, which models how reasoning and actions shape future task progress rather than simulating tool responses. AEWM combines \textbf{Action Judge} to distinguish \textsc{Critical}, \textsc{Exploratory}, and \textsc{Noisy} decisions with \textbf{State Revision} to edit noisy reasoning--action continuations from the same observed history. \textbf{EditAct} integrates these capabilities with real execution, directly changing the state underlying subsequent decisions rather than merely providing critiques. We train AEWM across Search, Terminal, and Software Engineering through mid-training and supervised fine-tuning. AEWM achieves 70.5\% macro-F1 on our Action Judge benchmark, exceeding the strongest frontier baseline by 10.6 points. Across six benchmarks and three agent backbones, EditAct improves average scores by 3.2--6.7 points over the strongest baseline. Furthermore, rejection sampling fine-tuning on verified EditAct trajectories, termed \textbf{AEWM-RFT}, improves over Self-RFT by 2.2--2.6 points across three domains without online AEWM guidance.

cs.CL↗
arXiv2026

Pareto-Guided Optimal Transport for Multi-Reward Alignment

Text-to-image generation models have achieved remarkable progress in preference optimization, yet achieving robust alignment across diverse reward models remains a significant challenge. Existing multi-reward fusion approaches rely on weighted summation, which is costly to tune and insufficient for balancing conflicting objectives. More critically, optimization with reward models is highly susceptible to reward hacking, where reward scores increase while the perceived quality of generated images deteriorates. We demonstrate that optimizing against a unified global target under heterogeneous reward upper bounds can induce reward hacking, a risk further exacerbated by the inherent instability of weak reward models. To mitigate this, we propose a Pareto Frontier-Guided Optimal Transport (PG-OT) framework. Our method constructs a prompt-specific Pareto frontier and maps dominated samples toward it via distribution-aware optimal transport. Furthermore, we develop both online and offline optimization strategies tailored to diverse reward signal characteristics. To provide a more rigorous assessment, we introduce the Joint Domination Rate (JDR) and Joint Collapse Rate (JCR) as principled metrics to quantify multi-reward synergy and reward hacking. Experimental results show that our approach outperforms strong baselines with an 11% gain in JDR and achieves a near 80% win rate in human evaluations.

cs.CV↗
arXiv2026

Rethinking the Evaluation and Optimization of LLM-Based Social Simulation

LLM-based social simulation is a promising complement to traditional methods such as surveys and behavioral experiments. A core question is how to evaluate the fidelity of LLM-simulated human behavior and optimize LLMs toward it. Prevailing practice evaluates by accuracy, checking whether the model selects the single response observed from a human, and trains the LLM to reproduce this hard label. However, human behavior is inherently subjective: the same person in the same situation may reasonably act differently, so an observed response is only one draw from an underlying response distribution, rendering accuracy-based evaluation unreliable and hard-label training misleading. To address these problems, we first introduce the subjectivity coefficient, an entropy-based quantity distinguishing objective tasks such as coding from subjective ones such as social simulation, and use it to systematically analyze how accuracy-based evaluation and hard-label training fail as subjectivity grows. Based on the subjectivity coefficient, we propose Subjectivity-Adaptive soft-Label Training (SALT): it pools observed outputs from semantically nearby inputs into soft distributional labels, with an aggregation radius adapted to the estimated subjectivity of each input; in the near-objective limit the neighborhood shrinks, so SALT naturally falls back to standard single-label training. Moreover, since existing datasets record only single observed responses and cannot support distributional evaluation, we construct SUBJSIM, a benchmark of 19,300 contexts covering 193 annotators and 100 subjective questions. Since real-world data typically provide only a single observation per input, our experiments train models from single observed outputs while evaluating them against the full response distributions, verifying feasibility in realistic settings. Results on SUBJSIM demonstrate the advantages of our method.

cs.AI↗
arXiv2026

ClawGym II: Exploring Black-Box RL on Agent Harness

Agent harnesses have substantially improved performance on long-horizon tasks by coordinating agent interactions with the environment. However, reinforcement learning through complex harnesses remains largely unexplored, as scaling such training to long-horizon agent tasks introduces fundamental challenges. In this work, we present a unified black-box RL framework for stable and scalable optimization of general agents through complex harnesses. Concretely, we first build a sandbox-based execution infrastructure that isolates task environments and harnesses within temporary sandboxes for large-scale concurrent rollouts. We then decouple policy optimization from opaque harness execution and place a serving proxy at the model boundary to capture model calls. To reconstruct multi-turn trajectories and improve training efficiency, we organize the captured calls into prefix trees and further adapt both critic-based PPO and critic-free GRPO to optimize over the recovered tree structure. Meanwhile, we maintain training-inference consistency throughout the optimization process. Finally, we introduce mix-harness training, allowing a single model to be jointly optimized by heterogeneous harnesses. With Qwen3-30A3B, black-box RL improves Pass@1 on ClawGym-Bench by 9.98 and 14.81 points through OpenClaw and Claude Code, respectively, while remaining stable over 200-400 optimization steps. Moreover, the framework yields consistent gains on more challenging tasks such as JobBench and OfficeQA. Overall, our framework enables effective, stable, and scalable optimization of general agents through black-box harnesses, supporting unified training across heterogeneous execution systems.

cs.CL↗
arXiv2026

LLaDA MoE v2: Scaling Mixture-of-Experts Diffusion Language Models

Diffusion language models (dLLMs) offer an alternative to autoregressive (AR) language modeling, yet the scaling behavior of Mixture-of-Experts (MoE) dLLMs remains poorly understood. We systematically characterize how optimization hyperparameters, compute allocation, and architecture scale for MoE dLLMs, identifying quantitative differences from scaling trends previously reported for AR models. Specifically, for optimization, the optimal nominal batch size grows faster, while the optimal learning rate decays more rapidly with compute. For model--data allocation, IsoFLOP analysis reveals a slight data-side tilt: the optimal token budget grows faster than activated model-side computation. For MoE architecture, larger scales increasingly favor larger expert pools at fixed activated capacity, while moderate expert granularity remains consistently effective and the preferred fraction of activated capacity assigned to shared experts remains stable across scales. Guided by these findings, we train LLaDA MoE v2, a 30B-A3B dLLM, from scratch on 23.5T tokens. With approximately 65\% as many pretraining tokens as Qwen3, LLaDA MoE v2 approaches Qwen3 on several knowledge, reasoning, and coding benchmarks. After supervised fine-tuning alone, it outperforms SDAR Chat on seven of eight reasoning and coding benchmarks and remains close to Qwen3 on several tasks. These results establish practical scaling laws and design principles for MoE dLLMs.

cs.AI↗
arXiv2026

AMTFV: Agentic Mathematical Tool-Flow Verification for LLM Self-Correction

Large language models have demonstrated strong mathematical problem-solving capabilities, yet reliably verifying their candidate answers remains challenging. Existing representative methods mainly revise outputs through natural-language reflection or assist verification by directly generating verification programs; the former may not reliably support exact computation, whereas the latter prematurely couples mathematical modeling with low-level implementation. We propose AMTFV (Agentic Mathematical Tool-Flow Verification). By introducing Mathematical Tool Flow (MTF) as an interrupt--execute--resume interface, AMTFV decouples verification modeling from concrete execution and supports exact computation through a mathematical toolbox. Specifically, the verification agent first constructs a verification workflow, encodes the mathematical objects and computational intent requiring reliable execution in an MTF request, and sends it to the mathematical toolbox agent. The latter parses the request, generates executable calls, and dispatches them to the backend for exact computation. Tool outputs then support candidate-answer adjudication, answer revision, and verification-workflow revision. We evaluate AMTFV on five challenging mathematical reasoning datasets with seven model configurations from DeepSeek, GPT, and Gemini. Experimental results show that AMTFV outperforms the representative baselines evaluated in this study overall; under an individual model configuration, it improves average accuracy over the strongest baseline by up to 8.3 percentage points, with larger gains on samples of medium and high verification complexity.

cs.AI↗