arXiv subjects

Mingyu Ma

Publications and source records attributed to Mingyu Ma.

6 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

Combating Instruction Conflict via Energy-Driven Latent Conflict Detection

Large Language Models (LLMs) are increasingly deployed with hierarchical instructions, yet they remain vulnerable to conflicts in which user directives override system-level constraints. Existing defense mechanisms predominantly focus on static input inspection and therefore fail to detect Response Drift, a phenomenon in which the model's final response violates system-level constraints despite seemingly compliant inputs. To bridge this gap, we introduce ELCD, a response-level latent conflict detector for post-generation, pre-delivery verification. Given the full generated output, ELCD constructs a composite hidden-state representation by concatenating the final-token embedding with the mean-pooled response embedding. It then optimizes a pairwise margin ranking objective to separate compliant and drifting responses in latent space. Extensive experiments across five mainstream LLMs ranging from 1.5B to 14B parameters demonstrate that ELCD significantly outperforms competitive baselines. Notably, it improves the PR-AUC on Llama-2-7B by approximately 30 percentage points and reduces the False Positive Rate at 95% TPR (FPR95) on Mistral-7B to 2.67%. These results suggest that ELCD provides a promising approach for latent instruction-conflict detection in open-weight or self-hosted LLM deployments.

cs.CL↗

arXiv2025

P4sim: Programming Protocol-independent Packet Processors in ns-3

Programmable data planes enable users to design data plane algorithms for network devices, providing extensive flexibility for network customization. Programming Protocol-Independent Packet Processors (P4) has become the most widely adopted abstraction, programming language, and framework for data plane programming. However, existing simulation platforms lack high-performance support for P4-based networks. This paper introduces P4sim, a high-performance P4-driven simulation framework built on bmv2 and NS4, seamlessly integrated with ns-3. It improves queue modeling, time scheduling, and P4 architecture support, extending compatibility to V1model, PSA, and PNA. P4sim enables efficient packet processing, accurate time tracking, and seamless interaction between P4-enabled hosts and switches. We evaluate the P4sim in terms of performance and queue management and demonstrate its capabilities using two common use cases: Basic Tunneling and Load Balancing. The results highlight the P4sim as a powerful tool for advancing research and education in programmable networks.

cs.NI↗

arXiv2018

Dual Memory Network Model for Biased Product Review Classification

In sentiment analysis (SA) of product reviews, both user and product information are proven to be useful. Current tasks handle user profile and product information in a unified model which may not be able to learn salient features of users and products effectively. In this work, we propose a dual user and product memory network (DUPMN) model to learn user profiles and product reviews using separate memory networks. Then, the two representations are used jointly for sentiment prediction. The use of separate models aims to capture user profiles and product information more effectively. Compared to state-of-the-art unified prediction models, the evaluations on three benchmark datasets, IMDB, Yelp13, and Yelp14, show that our dual learning model gives performance gain of 0.6%, 1.2%, and 0.9%, respectively. The improvements are also deemed very significant measured by p-values.

cs.CL↗

arXiv2017

Optimal double resonant condition in metallic core-shell nanocavity for third harmonic generation

As the rapid development of nonlinear optics, the enhancement of optical third harmonic generation becomes a pop research realm in the fields of physics, chemistry, biology, materials science, information science and other fields. In this letter, theoretical analysis on double resonance situation in metallic core-shell nanostructure is performed in order to optimize the efficiency of third harmonic generation. As plenty of cavity modes ranging from visible area to near infrared can be excited efficiently in core-shell nanospheres, kinds of double resonant conditions can be formed by matching two different multipolar cavity modes. Numerical simulations show that the third harmonic generation (THG) intensity in the far field can be enhanced remarkably when the THG signal couples well with the high order cavity mode. More importantly, THG efficiency is optimum from the cavity on double resonant conditions coupling two modes with the same order. The THG intensity differs up to 3 magnitudes on this condition. Subsequent theoretical analysis indicates that changing third-order nonlinear susceptibility of the metal shell while keeping that of the core fixed has almost no effect on THG efficiency. This finding about optimal double resonance has an appreciable effect on optimizing THG in spherical cavity structures.

physics.optics↗