Anjie Liu

Anjie Liu, Yan Song, Zhixun Chen et al.

Tool-augmented vision-language agents can acquire external perceptual evidence through OCR, detection, segmentation, and other tools, but executing every proposed tool call is costly and sometimes unnecessary. We study the pre-call control problem: after a ReAct-style VLM agent proposes a perceptual tool call, should the call be executed, or skipped before its output enters the context? Across five benchmarks, we find that the baseline agent exhibits poor local selectivity: helpful and harmful calls occur at similar rates (11.8% vs. 9.9%), while most calls do not change the immediate forced-answer prediction. We introduce ToolGate, a lightweight external controller that predicts execute/skip decisions from trajectory text and simple structural features. Across two Qwen3-VL backbones, ToolGate reduces token cost to 64-69% of the unrestricted ReAct baseline while preserving average accuracy in cross-domain settings. With matched-domain trajectory training on Qwen3-VL-30B, it further improves average accuracy by 1.65 points. These results show that tool-augmented VLM agents benefit not only from better perceptual tools, but also from explicit control over when tool outputs are worth paying for.

Huichi Zhou, Siyuan Guo, Anjie Liu et al.

We introduce \emph{Memento-Skills}, a generalist, continually-learnable LLM agent system that functions as an \emph{agent-designing agent}: it autonomously constructs, adapts, and improves task-specific agents through experience. The system is built on a memory-based reinforcement learning framework with \emph{stateful prompts}, where reusable skills (stored as structured markdown files) serve as persistent, evolving memory. These skills encode both behaviour and context, enabling the agent to carry forward knowledge across interactions. Starting from simple elementary skills (like Web search and terminal operations), the agent continually improves via the \emph{Read--Write Reflective Learning} mechanism introduced in \emph{Memento~2}~\cite{wang2025memento2}. In the \emph{read} phase, a behaviour-trainable skill router selects the most relevant skill conditioned on the current stateful prompt; in the \emph{write} phase, the agent updates and expands its skill library based on new experience. This closed-loop design enables \emph{continual learning without updating LLM parameters}, as all adaptation is realised through the evolution of externalised skills and prompts. Unlike prior approaches that rely on human-designed agents, Memento-Skills enables a generalist agent to \emph{design agents end-to-end} for new tasks. Through iterative skill generation and refinement, the system progressively improves its own capabilities. Experiments on the \emph{General AI Assistants} benchmark and \emph{Humanity's Last Exam} demonstrate sustained gains, achieving 26.2\% and 116.2\% relative improvements in overall accuracy, respectively. Code is available at https://github.com/Memento-Teams/Memento-Skills.

Jun Wang, Meng Fang, Ziyu Wan et al.

In this technical report, we introduce OpenR, an open-source framework designed to integrate key components for enhancing the reasoning capabilities of large language models (LLMs). OpenR unifies data acquisition, reinforcement learning training (both online and offline), and non-autoregressive decoding into a cohesive software platform. Our goal is to establish an open-source platform and community to accelerate the development of LLM reasoning. Inspired by the success of OpenAI's o1 model, which demonstrated improved reasoning abilities through step-by-step reasoning and reinforcement learning, OpenR integrates test-time compute, reinforcement learning, and process supervision to improve reasoning in LLMs. Our work is the first to provide an open-source framework that explores the core techniques of OpenAI's o1 model with reinforcement learning, achieving advanced reasoning capabilities beyond traditional autoregressive methods. We demonstrate the efficacy of OpenR by evaluating it on the MATH dataset, utilising publicly available data and search methods. Our initial experiments confirm substantial gains, with relative improvements in reasoning and performance driven by test-time computation and reinforcement learning through process reward models. The OpenR framework, including code, models, and datasets, is accessible at https://openreasoner.github.io.

Anjie Liu, Ziqin Gong, Yan Song et al.

Visual perception in modern Vision-Language Models (VLMs) is constrained by a fundamental perceptual bandwidth bottleneck: a broad field of view inevitably sacrifices the fine-grained details necessary for complex reasoning. Inspired by the classical paradigms of active vision and information foraging, we frame overcoming this limitation as a sequential decision-making process. We formalise this process through the lens of the sequential Bayesian optimal experimental design (S-BOED) problem. While exact Bayesian inference is intractable in continuous gigapixel spaces, we derive principled yet tractable approximations that balance spatial coverage against resolution. To validate this framework, we present a training-free inference strategy as a practical instantiation of the S-BOED objective for agents equipped with multiple vision tools. Designed as a flexible template, this strategy accommodates arbitrary optimisation algorithms, ranging from efficient greedy sampling to look-ahead planning, to approximate the optimal design. Empirical evaluations on gigapixel-level benchmarks demonstrate that our approach further boosts the performance of state-of-the-art models, significantly outperforming standard baselines and effectively narrowing the gap towards human-annotated oracles.

Xiangning Yu, Zhuohan Wang, Linyi Yang et al. · pku

Chain-of-Thought (CoT) prompting plays an indispensable role in endowing large language models (LLMs) with complex reasoning capabilities. However, CoT currently faces two fundamental challenges: (1) Sufficiency, which ensures that the generated intermediate inference steps comprehensively cover and substantiate the final conclusion; and (2) Necessity, which identifies the inference steps that are truly indispensable for the soundness of the resulting answer. We propose a causal framework that characterizes CoT reasoning through the dual lenses of sufficiency and necessity. Incorporating causal Probability of Sufficiency and Necessity allows us not only to determine which steps are logically sufficient or necessary to the prediction outcome, but also to quantify their actual influence on the final reasoning outcome under different intervention scenarios, thereby enabling the automated addition of missing steps and the pruning of redundant ones. Extensive experimental results on various mathematical and commonsense reasoning benchmarks confirm substantial improvements in reasoning efficiency and reduced token usage without sacrificing accuracy. Our work provides a promising direction for improving LLM reasoning performance and cost-effectiveness.

Anjie Liu, Jianhong Wang, Samuel Kaski et al.

Steering cooperative multi-agent reinforcement learning (MARL) towards desired outcomes is challenging, particularly when the global guidance from a human on the whole multi-agent system is impractical in a large-scale MARL. On the other hand, designing external mechanisms (e.g., intrinsic rewards and human feedback) to coordinate agents mostly relies on empirical studies, lacking a easy-to-use research tool. In this work, we employ multi-agent influence diagrams (MAIDs) as a graphical framework to address the above issues. First, we introduce the concept of MARL interaction paradigms (orthogonal to MARL learning paradigms), using MAIDs to analyze and visualize both unguided self-organization and global guidance mechanisms in MARL. Then, we design a new MARL interaction paradigm, referred to as the targeted intervention paradigm that is applied to only a single targeted agent, so the problem of global guidance can be mitigated. In implementation, we introduce a causal inference technique, referred to as Pre-Strategy Intervention (PSI), to realize the targeted intervention paradigm. Since MAIDs can be regarded as a special class of causal diagrams, a composite desired outcome that integrates the primary task goal and an additional desired outcome can be achieved by maximizing the corresponding causal effect through the PSI. Moreover, the bundled relevance graph analysis of MAIDs provides a tool to identify whether an MARL learning paradigm is workable under the design of an MARL interaction paradigm. In experiments, we demonstrate the effectiveness of our proposed targeted intervention, and verify the result of relevance graph analysis.

Anjie Liu, Jinglang W. Sun, Anh Ngo et al.

Establishing accurate field development parameters to optimize long-term oil production takes time and effort due to the complexity of oil well development, and the uncertainty in estimating long-term well production. Traditionally, oil and gas companies use simulation software that are inherently computationally expensive to forecast production. Thus, machine learning approaches are recently utilized in literature as an efficient alternative to optimize well developments by enhancing completion conditions. The primary goal of this project is to develop effective machine-learning models that can integrate the effects of multidimensional predictive variables (i.e., completion conditions) to predict 12-Month Cumulative Production accurately. Three predictive regression machine learning models are implemented for predicting 12-month cumulative oil production: Random Forest, Gradient Boosting, and Long Short-Term Memory Models. All three models yielded cumulative production predictions with root mean squared error (RMSE ) values ranging from 7.35 to 20.01 thousand barrels of oil. Although we hypothesized that all models would yield accurate predictions, the results indicated a crucial need for further refinement to create reliable and rational predictive tools in the subsurface. While this study did not produce optimal models for completion sequencing to maximize long-term production, we established that machine learning models alone are not self-sufficient for problems of this nature. Hence, there is potential for significant improvement, including comprehensive feature engineering, and a recommendation of exploring the use of hybrid or surrogate models (i.e., coupling physics reduced models and machine learning models), to ascertain significant contribution to the progress of completion sequencing workflows.