Search

Let $K$ be a finite extension of $\mathbb{Q}_p$, and $\rho$ be an $n$-dimensional (non-critical generic) crystabelline representation of the absolute Galois group of $K$ of regular Hodge-Tate weights. We associate to $\rho$ an explicit locally $\mathbb{Q}_p$-analytic representation $\pi_1(\rho)$ of $\mathrm{GL}_n(K)$, which encodes some $p$-adic Hodge parameters of $\rho$. When $K=\mathbb{Q}_p$, it encodes the full information hence reciprocally determines $\rho$. When $\rho$ is associated to $p$-adic automorphic representations, we show under mild hypotheses that $\pi_1(\rho)$ is a subrepresentation of the $\mathrm{GL}_n(K)$-representation globally associated to $\rho$., Comment: 56 pages

Building generalist agents that can handle diverse tasks and evolve themselves across different environments is a long-term goal in the AI community. Large language models (LLMs) are considered a promising foundation to build such agents due to their generalized capabilities. Current approaches either have LLM-based agents imitate expert-provided trajectories step-by-step, requiring human supervision, which is hard to scale and limits environmental exploration; or they let agents explore and learn in isolated environments, resulting in specialist agents with limited generalization. In this paper, we take the first step towards building generally-capable LLM-based agents with self-evolution ability. We identify a trinity of ingredients: 1) diverse environments for agent exploration and learning, 2) a trajectory set to equip agents with basic capabilities and prior knowledge, and 3) an effective and scalable evolution method. We propose AgentGym, a new framework featuring a variety of environments and tasks for broad, real-time, uni-format, and concurrent agent exploration. AgentGym also includes a database with expanded instructions, a benchmark suite, and high-quality trajectories across environments. Next, we propose a novel method, AgentEvol, to investigate the potential of agent self-evolution beyond previously seen data across tasks and environments. Experimental results show that the evolved agents can achieve results comparable to SOTA models. We release the AgentGym suite, including the platform, dataset, benchmark, checkpoints, and algorithm implementations. The AgentGym suite is available on https://github.com/WooooDyy/AgentGym., Comment: Project site: https://agentgym.github.io

Let $K$ be a finite extension of $\mathbb{Q}_p$. We study the locally $\mathbb{Q}_p$-analytic representations $\pi$ of $\mathrm{GL}_n(K)$ of integral weights that appear in spaces of $p$-adic automorphic representations. We conjecture that the translation of $\pi$ to the singular block has an internal structure which is compatible with certain algebraic representations of $\mathrm{GL}_n$, analogously to the mod $p$ local-global compatibility conjecture of Breuil-Herzig-Hu-Morra-Schraen. We next make some conjectures and speculations on the wall-crossings of $\pi$. In particular, when $\pi$ is associated to a two dimensional de Rham Galois representation, we make conjectures and speculations on the relation between the Hodge filtrations of $\rho$ and the wall-crossings of $\pi$, which have a flavour of the Breuil-Strauch conjecture. We collect some results towards the conjectures and speculations., Comment: 82 pages, with an appendix by Yiwen Ding, Yongquan Hu and Haoran Wang

In this paper, we introduce the simulations and bisimulations on polarity-based semantics for non-distributive modal logic, which are natural generalizations of those notions on Kripke semantics for modal logic. We also generalize other important model-theoretic notions about Kripke semantics such as image-finite models, modally-saturated models, ultrafilter extension and ultrapower extension to the non-distributive setting. By using these generalizations, we prove the Hennessy-Milner theorem and the van Benthem characterization theorem for non-distributive modal logic based on polarity-based semantics.

Large language models (LLMs) have shown promising abilities of in-context learning (ICL), adapting swiftly to new tasks with only few-shot demonstrations. However, current few-shot methods heavily depend on high-quality, query-specific demos, which are often lacking. When faced with out-of-demonstration (OOD) queries, methods that rely on hand-crafted demos or external retrievers might fail. To bridge the gap between limited demos and OOD queries, we propose Self-Demos, a novel prompting method that elicits the inherent generalizability in LLMs by query-aware demo generation. The generated demos strategically interpolate between existing demos and the given query, transforming the query from OOD to ID. To evaluate the effectiveness of our approach, we manually constructed OOD-Toolset, a dataset in the tool-using scenario with over 300 real-world APIs and 1000 instances, each consisting of three tool-use cases as demos and an OOD query. Thorough experiments on our dataset and two public math benchmarks have shown that our method can outperform state-of-the-art baselines in the OOD setting. Moreover, we conduct a range of analyses to validate Self-Demos's generalization and provide more insights., Comment: Accepted to NAACL 2024 Findings

Large language models (LLMs) have demonstrated impressive performance in understanding language and executing complex reasoning tasks. However, LLMs with long context windows have been notorious for their expensive training costs and high inference latency. Even the most advanced models such as GPT-4 and Claude2 often make mistakes when processing inputs of over $100k$ tokens, a phenomenon also known as \textit{lost in the middle}. In this paper, we propose \textsc{LongAgent}, a method based on multi-agent collaboration, which scales LLMs (e.g., LLaMA) to a context of 128K and demonstrates potential superiority in long-text processing compared to GPT-4. In \textsc{LongAgent}, a leader is responsible for understanding user intent and directing team members to acquire information from documents. Due to members' hallucinations, it is non-trivial for a leader to obtain accurate information from the responses of dozens to hundreds of members. To address this, we develop an \textit{inter-member communication} mechanism to resolve response conflicts caused by hallucinations through information sharing. Our experimental results indicate that \textsc{LongAgent} offers a promising alternative for long-text processing. The agent team instantiated with LLaMA-7B achieves significant improvements in tasks such as 128k-long text retrieval, multi-hop question answering, compared to GPT-4.

In this paper, we propose R$^3$: Learning Reasoning through Reverse Curriculum Reinforcement Learning (RL), a novel method that employs only outcome supervision to achieve the benefits of process supervision for large language models. The core challenge in applying RL to complex reasoning is to identify a sequence of actions that result in positive rewards and provide appropriate supervision for optimization. Outcome supervision provides sparse rewards for final results without identifying error locations, whereas process supervision offers step-wise rewards but requires extensive manual annotation. R$^3$ overcomes these limitations by learning from correct demonstrations. Specifically, R$^3$ progressively slides the start state of reasoning from a demonstration's end to its beginning, facilitating easier model exploration at all stages. Thus, R$^3$ establishes a step-wise curriculum, allowing outcome supervision to offer step-level signals and precisely pinpoint errors. Using Llama2-7B, our method surpasses RL baseline on eight reasoning tasks by $4.1$ points on average. Notebaly, in program-based reasoning on GSM8K, it exceeds the baseline by $4.2$ points across three backbone models, and without any extra data, Codellama-7B + R$^3$ performs comparable to larger models or closed-source models., Comment: Preprint. Codes released: https://github.com/WooooDyy/LLM-Reverse-Curriculum-RL

Cereal grain plays a crucial role in the human diet as a major source of essential nutrients. Grain Appearance Inspection (GAI) serves as an essential process to determine grain quality and facilitate grain circulation and processing. However, GAI is routinely performed manually by inspectors with cumbersome procedures, which poses a significant bottleneck in smart agriculture. In this paper, we endeavor to develop an automated GAI system:AI4GrainInsp. By analyzing the distinctive characteristics of grain kernels, we formulate GAI as a ubiquitous problem: Anomaly Detection (AD), in which healthy and edible kernels are considered normal samples while damaged grains or unknown objects are regarded as anomalies. We further propose an AD model, called AD-GAI, which is trained using only normal samples yet can identify anomalies during inference. Moreover, we customize a prototype device for data acquisition and create a large-scale dataset including 220K high-quality images of wheat and maize kernels. Through extensive experiments, AD-GAI achieves considerable performance in comparison with advanced AD methods, and AI4GrainInsp has highly consistent performance compared to human experts and excels at inspection efficiency over 20x speedup. The dataset, code and models will be released at https://github.com/hellodfan/AI4GrainInsp., Comment: Accepted by ECAI2023. https://github.com/hellodfan/AI4GrainInsp

In this paper, we generalize modal $\mu$-calculus to the non-distributive (lattice-based) modal $\mu$-calculus and formalize some scenarios regarding categorization using it. We also provide a game semantics for the developed logic. The proof of adequacy of this game semantics proceeds by generalizing the unfolding games on the power-set algebras to the arbitrary lattices and showing that these games can be used to determine the least and the greatest fixed points of a monotone operator on a lattice. Finally, we define a notion of bisimulations on the polarities and show invariance of non-distributive modal $\mu$-calculus under them.

For a long time, humanity has pursued artificial intelligence (AI) equivalent to or surpassing the human level, with AI agents considered a promising vehicle for this pursuit. AI agents are artificial entities that sense their environment, make decisions, and take actions. Many efforts have been made to develop intelligent agents, but they mainly focus on advancement in algorithms or training strategies to enhance specific capabilities or performance on particular tasks. Actually, what the community lacks is a general and powerful model to serve as a starting point for designing AI agents that can adapt to diverse scenarios. Due to the versatile capabilities they demonstrate, large language models (LLMs) are regarded as potential sparks for Artificial General Intelligence (AGI), offering hope for building general AI agents. Many researchers have leveraged LLMs as the foundation to build AI agents and have achieved significant progress. In this paper, we perform a comprehensive survey on LLM-based agents. We start by tracing the concept of agents from its philosophical origins to its development in AI, and explain why LLMs are suitable foundations for agents. Building upon this, we present a general framework for LLM-based agents, comprising three main components: brain, perception, and action, and the framework can be tailored for different applications. Subsequently, we explore the extensive applications of LLM-based agents in three aspects: single-agent scenarios, multi-agent scenarios, and human-agent cooperation. Following this, we delve into agent societies, exploring the behavior and personality of LLM-based agents, the social phenomena that emerge from an agent society, and the insights they offer for human society. Finally, we discuss several key topics and open problems within the field. A repository for the related papers at https://github.com/WooooDyy/LLM-Agent-Paper-List., Comment: 86 pages, 12 figures

This work extends Halpern and Pearl's causal models for actual causality to a possible world semantics environment. Using this framework we introduce a logic of actual causality with modal operators, which allows for reasoning about causality in scenarios involving multiple possibilities, temporality, knowledge and uncertainty. We illustrate this with a number of examples, and conclude by discussing some future directions for research., Comment: In Proceedings TARK 2023, arXiv:2307.04005

Let $D_1\subset D_2$ be $(\varphi, \Gamma)$-modules of rank $2$ over the Robba ring, and $\pi(D_1)$, $\pi(D_2)$ be the associated locally analytic representations of $\rm{GL}_2(\mathbb{Q}_p)$ via the $p$-adic local Langlands correspondence. We describe the relation between $\pi(D_1)$ and $\pi(D_2)$., Comment: 19 pages

We prove Breuil's conjecture on locally analytic $\rm{Ext}^1$ for $\rm{GL}_2(\mathbb{Q}_p)$ in de Rham non-trianguline case., Comment: 12 pages

While several previous studies have analyzed gender bias in research, we are still missing a comprehensive analysis of gender differences in the AI community, covering diverse topics and different development trends. Using the AI Scholar dataset of 78K researchers in the field of AI, we identify several gender differences: (1) Although female researchers tend to have fewer overall citations than males, this citation difference does not hold for all academic-age groups; (2) There exist large gender homophily in co-authorship on AI papers; (3) Female first-authored papers show distinct linguistic styles, such as longer text, more positive emotion words, and more catchy titles than male first-authored papers. Our analysis provides a window into the current demographic trends in our AI community, and encourages more gender equality and diversity in the future. Our code and data are at https://github.com/causalNLP/ai-scholar-gender.

Cereal grains are a vital part of human diets and are important commodities for people's livelihood and international trade. Grain Appearance Inspection (GAI) serves as one of the crucial steps for the determination of grain quality and grain stratification for proper circulation, storage and food processing, etc. GAI is routinely performed manually by qualified inspectors with the aid of some hand tools. Automated GAI has the benefit of greatly assisting inspectors with their jobs but has been limited due to the lack of datasets and clear definitions of the tasks. In this paper we formulate GAI as three ubiquitous computer vision tasks: fine-grained recognition, domain adaptation and out-of-distribution recognition. We present a large-scale and publicly available cereal grains dataset called GrainSpace. Specifically, we construct three types of device prototypes for data acquisition, and a total of 5.25 million images determined by professional inspectors. The grain samples including wheat, maize and rice are collected from five countries and more than 30 regions. We also develop a comprehensive benchmark based on semi-supervised learning and self-supervised learning techniques. To the best of our knowledge, GrainSpace is the first publicly released dataset for cereal grain inspection., Comment: 8pages, 6 figures, accepted by CVPR2022 dataset is available at \url{https://github.com/hellodfan/GrainSpace}

Reasoning is central to human intelligence. However, fallacious arguments are common, and some exacerbate problems such as spreading misinformation about climate change. In this paper, we propose the task of logical fallacy detection, and provide a new dataset (Logic) of logical fallacies generally found in text, together with an additional challenge set for detecting logical fallacies in climate change claims (LogicClimate). Detecting logical fallacies is a hard problem as the model must understand the underlying logical structure of the argument. We find that existing pretrained large language models perform poorly on this task. In contrast, we show that a simple structure-aware classifier outperforms the best language model by 5.46% on Logic and 4.51% on LogicClimate. We encourage future work to explore this task as (a) it can serve as a new reasoning challenge for language models, and (b) it can have potential applications in tackling the spread of misinformation. Our dataset and code are available at https://github.com/causalNLP/logical-fallacy, Comment: EMNLP 2021 Findings

We construct parabolic analogues of (global) eigenvarieties, of patched eigenvarieties and of (local) trianguline varieties, that we call respectively Bernstein eigenvarieties, patched Bernstein eigenvarieties, and Bernstein paraboline varieties. We study the geometry of these rigid analytic spaces, in particular (generalizing results of Breuil-Hellmann-Schraen) we show that their local geometry can be described by certain algebraic schemes related to the generalized Grothendieck-Springer resolution. We deduce several local-global compatibility results, including a classicality result (with no trianguline assumption at $p$), and new cases towards the locally analytic socle conjecture of Breuil in the non-trianguline case., Comment: 149 pages

We study the adjunction property of the Jacquet-Emerton functor in certain neighborhoods of critical points in the eigencurve. As an application, we construct two-variable $p$-adic $L$-functions around critical points via Emerton's representation theoretic approach., Comment: 19 pages, to appear in Acta Mathematica Sinica, English Serie (Special Issue on Arithmetic Geometry)

We prove automorphy lifting results for certain essentially conjugate self-dual $p$-adic Galois representations $\rho$ over CM imaginary fields $F$, which satisfy in particular that $p$ splits in $F$, and that the restriction of $\rho$ on any decomposition group above $p$ is reducible with all the Jordan-H\"older factors of dimension at most $2$. We also show some results on Breuil's locally analytic socle conjecture in certain non-trianguline case. The main results are obtained by establishing an $R=\mathbb{T}$-type result over the $\mathrm{GL}_2(\mathbb{Q}_p)$-ordinary families considered by Breuil-Ding., Comment: 51 pages

We reinterpret a conjecture of Breuil on the locally analytic $\mathrm{Ext}^1$ in a functorial way using $(\varphi,\Gamma)$-modules (possibly with $t$-torsion) over the Robba ring, making it more accurate. Then we prove several special or partial cases of this "improved" conjecture, notably for ${\rm GL}_3(\mathbb{Q}_p)$., Comment: 157 pages, in French

Let $L$ be a finite extension of $\mathbb{Q}_p$, and $\rho_L$ be an $n$-dimensional semi-stable non crystalline $p$-adic representation of $\mathrm{Gal}_L$ with full monodromy rank. Via a study of Breuil's (simple) $\mathcal{L}$-invariants, we attach to $\rho_L$ a locally $\mathbb{Q}_p$-analytic representation $\Pi(\rho_L)$ of $\mathrm{GL}_n(L)$, which carries the exact information of the Fontaine-Mazur simple $\mathcal{L}$-invariants of $\rho_L$. When $\rho_L$ comes from an automorphic representation of $G(\mathbb{A}_{F^+})$ (for a unitary group $G$ over a totally real filed $F^+$ which is compact at infinite places and $\mathrm{GL}_n$ at $p$-adic places), we prove under mild hypothesis that $\Pi(\rho_L)$ is a subrerpresentation of the associated Hecke-isotypic subspaces of the Banach spaces of $p$-adic automorphic forms on $G(\mathbb{A}_{F^+})$. In other words, we prove the equality of Breuil's simple $\mathcal{L}$-invariants and Fontaine-Mazur simple $\mathcal{L}$-invariants., Comment: Final version

Let $\rho_p$ be a $3$-dimensional $p$-adic semi-stable representation of $\mathrm{Gal}(\overline{\mathbb{Q}_p}/\mathbb{Q}_p)$ with Hodge-Tate weights $(0,1,2)$ (up to shift) and such that $N^2\ne 0$ on $D_{\mathrm{st}}(\rho_p)$. When $\rho_p$ comes from an automorphic representation $\pi$ of $G(\mathbb{A}_{F^+})$ (for a unitary group $G$ over a totally real field $F^+$ which is compact at infinite places and $\mathrm{GL}_3$ at $p$-adic places), we show under mild genericity assumptions that the associated Hecke-isotypic subspaces of the Banach spaces of $p$-adic automorphic forms on $G(\mathbb{A}_{F^+}^\infty)$ of arbitrary fixed tame level contain (copies of) a unique admissible finite length locally analytic representation of $\mathrm{GL}_3(\mathbb{Q}_p)$ which only depends on and completely determines $\rho_p$., Comment: 139 pages