Large language models (LLMs) increasingly rely on multi-turn tool-integrated planning for knowledge-intensive and complex reasoning tasks. Existing implementations typically rely on a single agent, but they suffer from limited context length and noisy tool responses. A natural solution is to adopt a multi-agent framework with planner- and worker-agents to manage context. However, no existing methods support effective reinforcement learning post-training of tool-integrated multi-agent frameworks. To address this gap, we propose Multi-Agent Tool-Integrated Policy Optimization (MATPO), which enables distinct roles (planner and worker) to be trained within a single LLM instance using role-specific prompts via reinforcement learning. MATPO is derived from a principled credit assignment mechanism across planner and worker rollouts. This design eliminates the need to deploy multiple LLMs, which would be memory-intensive, while preserving the benefits of specialization. Experiments on GAIA-text, WebWalkerQA, and FRAMES show that MATPO consistently outperforms single-agent baselines by an average of 18.38% relative improvement in performance and exhibits greater robustness to noisy tool outputs. Our findings highlight the effectiveness of unifying multiple agent roles within a single LLM and provide practical insights for stable and efficient multi-agent RL training.

In an era where large language models (LLMs) are increasingly integrated into a wide range of everyday applications, research into these models' behavior has surged. However, due to the novelty of the field, clear methodological guidelines are lacking. This raises concerns about the replicability and generalizability of insights gained from research on LLM behavior. In this study, we discuss the potential risk of a replication crisis and support our concerns with a series of replication experiments focused on prompt engineering techniques purported to influence reasoning abilities in LLMs. We tested GPT-3.5, GPT-4o, Gemini 1.5 Pro, Claude 3 Opus, Llama 3-8B, and Llama 3-70B, on the chain-of-thought, EmotionPrompting, ExpertPrompting, Sandbagging, as well as Re-Reading prompt engineering techniques, using manually double-checked subsets of reasoning benchmarks including CommonsenseQA, CRT, NumGLUE, ScienceQA, and StrategyQA. Our findings reveal a general lack of statistically significant differences across nearly all techniques tested, highlighting, among others, several methodological weaknesses in previous research. We propose a forward-looking approach that includes developing robust methodologies for evaluating LLMs, establishing sound benchmarks, and designing rigorous experimental frameworks to ensure accurate and reliable assessments of model outputs.

The SuperCLUE-Fin (SC-Fin) benchmark is a pioneering evaluation framework tailored for Chinese-native financial large language models (FLMs). It assesses FLMs across six financial application domains and twenty-five specialized tasks, encompassing theoretical knowledge and practical applications such as compliance, risk management, and investment analysis. Using multi-turn, open-ended conversations that mimic real-life scenarios, SC-Fin measures models on a range of criteria, including accurate financial understanding, logical reasoning, clarity, computational efficiency, business acumen, risk perception, and compliance with Chinese regulations. In a rigorous evaluation involving over a thousand questions, SC-Fin identifies a performance hierarchy where domestic models like GLM-4 and MoonShot-v1-128k outperform others with an A-grade, highlighting the potential for further development in transforming theoretical knowledge into pragmatic financial solutions. This benchmark serves as a critical tool for refining FLMs in the Chinese context, directing improvements in financial knowledge databases, standardizing financial interpretations, and promoting models that prioritize compliance, risk management, and secure practices. We create a contextually relevant and comprehensive benchmark that drives the development of AI in the Chinese financial sector. SC-Fin facilitates the advancement and responsible deployment of FLMs, offering valuable insights for enhancing model performance and usability for both individual and institutional users in the Chinese market..~\footnote{Our benchmark can be found at \url{https://www.CLUEbenchmarks.com}}.

A reconstruction attack on a private dataset $D$ takes as input some publicly accessible information about the dataset and produces a list of candidate elements of $D$. We introduce a new class of data reconstruction attacks based on randomized methods for non-convex optimization. We empirically demonstrate that our attacks can not only reconstruct full rows of $D$ from aggregate query statistics $Q(D)\in \mathbb{R}^m$, but can do so in a way that reliably ranks reconstructed rows by their odds of appearing in the private data, providing a signature that could be used for prioritizing reconstructed rows for further actions such as identify theft or hate crime. We also design a sequence of baselines for evaluating reconstruction attacks. Our attacks significantly outperform those that are based only on access to a public distribution or population from which the private dataset $D$ was sampled, demonstrating that they are exploiting information in the aggregate statistics $Q(D)$, and not simply the overall structure of the distribution. In other words, the queries $Q(D)$ are permitting reconstruction of elements of this dataset, not the distribution from which $D$ was drawn. These findings are established both on 2010 U.S. decennial Census data and queries and Census-derived American Community Survey datasets. Taken together, our methods and experiments illustrate the risks in releasing numerically precise aggregate statistics of a large dataset, and provide further motivation for the careful application of provably private techniques such as differential privacy.