arXiv subjects

Andy Xu

Publications and source records attributed to Andy Xu.

9 recordsLinked to original sources

arXiv2026

Meta-Agent: From Task Descriptions to Verified Multi-Agent Systems

AI agents are increasingly used to solve complex, multi-step tasks, but existing multi-agent frameworks remain brittle as workflows grow in scale and depth. Small errors at intermediate stages can propagate through agent interactions, while insufficient grounding and weak verification mechanisms further limit reliability. We present Meta-Agent, a two-phase framework that automatically constructs and executes specialized multi-agent systems from natural-language task descriptions. In the construction phase, a task planner decomposes a problem into a directed acyclic graph of agent specifications with explicit input/output contracts and verification criteria. A web search module grounds each specification with external evidence, and a code generation module produces system prompts and tool configurations. A construction-time verification stage then validates generated artifacts and triggers targeted regeneration when failures are detected. In the execution phase, a coordinator dispatches subtasks across the agent graph while execution-time verification gates intermediate outputs. We further introduce a three-level error attribution mechanism that distinguishes local, upstream, and structural failures, enabling targeted recovery strategies ranging from localized retries to partial re-execution and re-decomposition. We evaluate Meta-Agent across coding, contextual learning, and open-ended reasoning tasks. Experiments against strong multi-agent baselines and ablation studies demonstrate consistent improvements in task success rate, error recovery, and workflow stability. The results highlight the importance of tightly integrating planning, grounding, and verification for building reliable multi-agent systems.

cs.AI↗

arXiv2025

LeMat-GenBench: A Unified Evaluation Framework for Crystal Generative Models

Generative machine learning (ML) models hold great promise for accelerating materials discovery through the inverse design of inorganic crystals, enabling an unprecedented exploration of chemical space. Yet, the lack of standardized evaluation frameworks makes it challenging to evaluate, compare, and further develop these ML models meaningfully. In this work, we introduce LeMat-GenBench, a unified benchmark for generative models of crystalline materials, supported by a set of evaluation metrics designed to better inform model development and downstream applications. We release both an open-source evaluation suite and a public leaderboard on Hugging Face, and benchmark 12 recent generative models. Results reveal that an increase in stability leads to a decrease in novelty and diversity on average, with no model excelling across all dimensions. Altogether, LeMat-GenBench establishes a reproducible and extensible foundation for fair model comparison and aims to guide the development of more reliable, discovery-oriented generative models for crystalline materials.

cs.LG↗

arXiv2025

LUQ: Layerwise Ultra-Low Bit Quantization for Multimodal Large Language Models

Large Language Models (LLMs) with multimodal capabilities have revolutionized vision-language tasks, but their deployment often requires huge memory and computational resources. Post-training quantization (PTQ) has successfully compressed language models to as low as 1-bit precision, its effectiveness for multimodal LLMs (MLLMs) remains unexplored. In this paper, we present the first method for ultra-low-bit (<4-bit) quantization of MLLMs. Our analysis reveals that multimodal tokens and intermediate layer activations produced by them exhibit significantly higher entropy compared to text tokens, indicating greater functional complexity that makes MLLMs less tolerant to ultra-low bit quantization. However, this entropy varies significantly across layers, with some layers producing lower-entropy activation distributions that we empirically show can better tolerate ultra-low bit quantization. Existing PTQ methods optimize weight quantization within each layer but apply the same target precision uniformly, ignoring this variation in complexity across layers. Building on this insight, we propose LUQ: Layerwise Ultra-Low Bit Quantization, which characterizes each transformer layer's functional complexity via its output activation entropy and selectively applies ultra-low bit quantization to layers encoding simpler, more compressible functions. We also show that multimodal calibration (image and text tokens) boosts VQA performance in the ultra-low bit regime. Evaluated on LLaVA-1.5 and Qwen-2.5-VL across 9 VQA benchmarks, LUQ models use 40% and 31% less memory than their 4-bit counterparts while exhibiting less than 10% degradation on MME.

cs.CV↗

arXiv2025

PLaID++: A Preference Aligned Language Model for Targeted Inorganic Materials Design

Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a promising approach to improve correctness in LLMs, however, in many scientific problems, the objective is not necessarily to produce the correct answer, but instead to produce a diverse array of candidates which satisfy a set of constraints. We study this challenge in the context of materials generation. To this end, we introduce PLaID++, an LLM post-trained for stable and property-guided crystal generation. We find that performance hinges on our crystallographic representation and reward formulation. First, we introduce a compact, symmetry-informed Wyckoff text representation which improves computational efficiency and encourages generalization from physical priors. Second, we demonstrate that temperature scaling acts as an entropy regularizer which counteracts mode collapse and encourages exploration. By encoding symmetry constraints directly into text and guiding model outputs towards desirable chemical space, PLaID++ generates structures that are thermodynamically stable, unique, and novel at a $\sim$50\% greater rate than prior methods and conditionally generates structures with desired space group properties. Our work demonstrates the potential of adapting post-training techniques from natural language processing to materials design, paving the way for targeted and efficient discovery of novel materials.

cs.LG↗

arXiv2025

SmartAvatar: Text- and Image-Guided Human Avatar Generation with VLM AI Agents

SmartAvatar is a vision-language-agent-driven framework for generating fully rigged, animation-ready 3D human avatars from a single photo or textual prompt. While diffusion-based methods have made progress in general 3D object generation, they continue to struggle with precise control over human identity, body shape, and animation readiness. In contrast, SmartAvatar leverages the commonsense reasoning capabilities of large vision-language models (VLMs) in combination with off-the-shelf parametric human generators to deliver high-quality, customizable avatars. A key innovation is an autonomous verification loop, where the agent renders draft avatars, evaluates facial similarity, anatomical plausibility, and prompt alignment, and iteratively adjusts generation parameters for convergence. This interactive, AI-guided refinement process promotes fine-grained control over both facial and body features, enabling users to iteratively refine their avatars via natural-language conversations. Unlike diffusion models that rely on static pre-trained datasets and offer limited flexibility, SmartAvatar brings users into the modeling loop and ensures continuous improvement through an LLM-driven procedural generation and verification system. The generated avatars are fully rigged and support pose manipulation with consistent identity and appearance, making them suitable for downstream animation and interactive applications. Quantitative benchmarks and user studies demonstrate that SmartAvatar outperforms recent text- and image-driven avatar generation systems in terms of reconstructed mesh quality, identity fidelity, attribute accuracy, and animation readiness, making it a versatile tool for realistic, customizable avatar creation on consumer-grade hardware.

cs.CV↗

arXiv2025

ad-trait: A Fast and Flexible Automatic Differentiation Library in Rust

The Rust programming language is an attractive choice for robotics and related fields, offering highly efficient and memory-safe code. However, a key limitation preventing its broader adoption in these domains is the lack of high-quality, well-supported Automatic Differentiation (AD)-a fundamental technique that enables convenient derivative computation by systematically accumulating data during function evaluation. In this work, we introduce ad-trait, a new Rust-based AD library. Our implementation overloads Rust's standard floating-point type with a flexible trait that can efficiently accumulate necessary information for derivative computation. The library supports both forward-mode and reverse-mode automatic differentiation, making it the first operator-overloading AD implementation in Rust to offer both options. Additionally, ad-trait leverages Rust's performance-oriented features, such as Single Instruction, Multiple Data acceleration in forward-mode AD, to enhance efficiency. Through benchmarking experiments, we show that our library is among the fastest AD implementations across several programming languages for computing derivatives. Moreover, it is already integrated into a Rust-based robotics library, where we showcase its ability to facilitate fast optimization procedures. We conclude with a discussion of the limitations and broader implications of our work.

cs.RO↗

arXiv2024

SoundnessBench: A Soundness Benchmark for Neural Network Verifiers

Neural network (NN) verification aims to formally verify properties of NNs, which is crucial for ensuring the behavior of NN-based models in safety-critical applications. In recent years, the community has developed many NN verifiers and benchmarks to evaluate them. However, existing benchmarks typically lack ground-truth for hard instances where no current verifier can verify the property and no counterexample can be found. This makes it difficult to validate the soundness of a verifier, when it claims verification on such challenging instances that no other verifier can handle. In this work, we develop a new benchmark for NN verification, named SoundnessBench, specifically for testing the soundness of NN verifiers. SoundnessBench consists of instances with deliberately inserted counterexamples that are hidden from adversarial attacks commonly used to find counterexamples. Thereby, it can identify false verification claims when hidden counterexamples are known to exist. We design a training method to produce NNs with hidden counterexamples and systematically construct our SoundnessBench with instances across various model architectures, activation functions, and input data. We demonstrate that our training effectively produces hidden counterexamples and our SoundnessBench successfully identifies bugs in state-of-the-art NN verifiers. Our code is available at https://github.com/mvp-harry/SoundnessBench and our dataset is available at https://huggingface.co/datasets/SoundnessBench/SoundnessBench.

cs.LG↗

arXiv2024

HELP: Hierarchical Embeddings-based Log Parsing

Logs are a first-hand source of information for software maintenance and failure diagnosis. Log parsing, which converts semi-structured log messages into structured templates, is a prerequisite for automated log analysis tasks such as anomaly detection, troubleshooting, and root cause analysis. However, existing log parsers fail in real-world systems for three main reasons. First, traditional heuristics-based parsers require handcrafted features and domain knowledge, which are difficult to generalize at scale. Second, existing large language model-based parsers rely on periodic offline processing, limiting their effectiveness in real-time use cases. Third, existing online parsing algorithms are susceptible to log drift, where slight log changes create false positives that drown out real anomalies. To address these challenges, we propose HELP, a Hierarchical Embeddings-based Log Parser. HELP is the first online semantic-based parser to leverage LLMs for performant and cost-effective log parsing. We achieve this through a novel hierarchical embeddings module, which fine-tunes a text embedding model to cluster logs before parsing, reducing querying costs by multiple orders of magnitude. To combat log drift, we also develop an iterative rebalancing module, which periodically updates existing log groupings. We evaluate HELP extensively on 14 public large-scale datasets, showing that HELP achieves significantly higher F1-weighted grouping and parsing accuracy than current state-of-the-art online log parsers. We also implement HELP into Iudex's production observability platform, confirming HELP's practicality in a production environment. Our results show that HELP is effective and efficient for high-throughput real-world log parsing.

cs.SE↗

arXiv2023

Optimal and Stable Multi-Layer Object Rearrangement on a Tabletop

Object rearrangement is a fundamental sub-task in accomplishing a great many physical tasks. As such, effectively executing rearrangement is an important skill for intelligent robots to master. In this study, we conduct the first algorithmic study on optimally solving the problem of Multi-layer Object Rearrangement on a Tabletop (MORT), in which one object may be relocated at a time, and an object can only be moved if other objects do not block its top surface. In addition, any intermediate structure during the reconfiguration process must be physically stable, i.e., it should stand without external support. To tackle the dual challenges of untangling the dependencies between objects and ensuring structural stability, we develop an algorithm that interleaves the computation of the optimal rearrangement plan and structural stability checking. Using a carefully constructed integer linear programming (ILP) model, our algorithm, Stability-aware Integer Programming-based Planner (SIPP), readily scales to optimally solve complex rearrangement problems of 3D structures with over 60 building blocks, with solution quality significantly outperforming natural greedy best-first approaches. Upon the publication of the manuscript, source code and data will be available at https://github.com/arc-l/mort/

cs.RO↗