ResearchPulse: Building Method-Experiment Chains through Multi-Document Scientific Inference

Understanding how scientific ideas evolve requires more than summarizing individual papers-it demands structured, cross-document reasoning over thematically related research. In this work, we formalize multi-document scientific inference, a new task that extracts and aligns motivation, methodology, and experimental results across related papers to reconstruct research development chains. This task introduces key challenges, including temporally aligning loosely structured methods and standardizing heterogeneous experimental tables. We present ResearchPulse, an agent-based framework that integrates instruction planning, scientific content extraction, and structured visualization. It consists of three coordinated agents: a Plan Agent for task decomposition, a Mmap-Agent that constructs motivation-method mind maps, and a Lchart-Agent that synthesizes experimental line charts. To support this task, we introduce ResearchPulse-Bench, a citation-aware benchmark of annotated paper clusters. Experiments show that our system, despite using 7B-scale agents, consistently outperforms strong baselines like GPT-4o in semantic alignment, structural consistency, and visual fidelity. The dataset are available in https://huggingface.co/datasets/ResearchPulse/ResearchPulse-Bench.

A Survey of Scientific Large Language Models: From Data Foundations to Agent Frontiers

Scientific Large Language Models (Sci-LLMs) are transforming how knowledge is represented, integrated, and applied in scientific research, yet their progress is shaped by the complex nature of scientific data. This survey presents a comprehensive, data-centric synthesis that reframes the development of Sci-LLMs as a co-evolution between models and their underlying data substrate. We formulate a unified taxonomy of scientific data and a hierarchical model of scientific knowledge, emphasizing the multimodal, cross-scale, and domain-specific challenges that differentiate scientific corpora from general natural language processing datasets. We systematically review recent Sci-LLMs, from general-purpose foundations to specialized models across diverse scientific disciplines, alongside an extensive analysis of over 270 pre-/post-training datasets, showing why Sci-LLMs pose distinct demands -- heterogeneous, multi-scale, uncertainty-laden corpora that require representations preserving domain invariance and enabling cross-modal reasoning. On evaluation, we examine over 190 benchmark datasets and trace a shift from static exams toward process- and discovery-oriented assessments with advanced evaluation protocols. These data-centric analyses highlight persistent issues in scientific data development and discuss emerging solutions involving semi-automated annotation pipelines and expert validation. Finally, we outline a paradigm shift toward closed-loop systems where autonomous agents based on Sci-LLMs actively experiment, validate, and contribute to a living, evolving knowledge base. Collectively, this work provides a roadmap for building trustworthy, continually evolving artificial intelligence (AI) systems that function as a true partner in accelerating scientific discovery.

StructVRM: Aligning Multimodal Reasoning with Structured and Verifiable Reward Models

Existing Vision-Language Models often struggle with complex, multi-question reasoning tasks where partial correctness is crucial for effective learning. Traditional reward mechanisms, which provide a single binary score for an entire response, are too coarse to guide models through intricate problems with multiple sub-parts. To address this, we introduce StructVRM, a method that aligns multimodal reasoning with Structured and Verifiable Reward Models. At its core is a model-based verifier trained to provide fine-grained, sub-question-level feedback, assessing semantic and mathematical equivalence rather than relying on rigid string matching. This allows for nuanced, partial credit scoring in previously intractable problem formats. Extensive experiments demonstrate the effectiveness of StructVRM. Our trained model, Seed-StructVRM, achieves state-of-the-art performance on six out of twelve public multimodal benchmarks and our newly curated, high-difficulty STEM-Bench. The success of StructVRM validates that training with structured, verifiable rewards is a highly effective approach for advancing the capabilities of multimodal models in complex, real-world reasoning domains.

Magentic-UI: Towards Human-in-the-loop Agentic Systems

AI agents powered by large language models are increasingly capable of autonomously completing complex, multi-step tasks using external tools. Yet, they still fall short of human-level performance in most domains including computer use, software development, and research. Their growing autonomy and ability to interact with the outside world, also introduces safety and security risks including potentially misaligned actions and adversarial manipulation. We argue that human-in-the-loop agentic systems offer a promising path forward, combining human oversight and control with AI efficiency to unlock productivity from imperfect systems. We introduce Magentic-UI, an open-source web interface for developing and studying human-agent interaction. Built on a flexible multi-agent architecture, Magentic-UI supports web browsing, code execution, and file manipulation, and can be extended with diverse tools via Model Context Protocol (MCP). Moreover, Magentic-UI presents six interaction mechanisms for enabling effective, low-cost human involvement: co-planning, co-tasking, multi-tasking, action guards, and long-term memory. We evaluate Magentic-UI across four dimensions: autonomous task completion on agentic benchmarks, simulated user testing of its interaction capabilities, qualitative studies with real users, and targeted safety assessments. Our findings highlight Magentic-UI's potential to advance safe and efficient human-agent collaboration.

Protein-SE(3): Benchmarking SE(3)-based Generative Models for Protein Structure Design

SE(3)-based generative models have shown great promise in protein geometry modeling and effective structure design. However, the field currently lacks a modularized benchmark to enable comprehensive investigation and fair comparison of different methods. In this paper, we propose Protein-SE(3), a new benchmark based on a unified training framework, which comprises protein scaffolding tasks, integrated generative models, high-level mathematical abstraction, and diverse evaluation metrics. Recent advanced generative models designed for protein scaffolding, from multiple perspectives like DDPM (Genie1 and Genie2), Score Matching (FrameDiff and RfDiffusion) and Flow Matching (FoldFlow and FrameFlow) are integrated into our framework. All integrated methods are fairly investigated with the same training dataset and evaluation metrics. Furthermore, we provide a high-level abstraction of the mathematical foundations behind the generative models, enabling fast prototyping of future algorithms without reliance on explicit protein structures. Accordingly, we release the first comprehensive benchmark built upon unified training framework for SE(3)-based protein structure design, which is publicly accessible at https://github.com/BruthYU/protein-se3.

AlphaFold Database Debiasing for Robust Inverse Folding

The AlphaFold Protein Structure Database (AFDB) offers unparalleled structural coverage at near-experimental accuracy, positioning it as a valuable resource for data-driven protein design. However, its direct use in training deep models that are sensitive to fine-grained atomic geometry, such as inverse folding, exposes a critical limitation. Comparative analysis of structural feature distributions reveals that AFDB structures exhibit distinct statistical regularities, reflecting a systematic geometric bias that deviates from the conformational diversity found in experimentally determined structures from the Protein Data Bank (PDB). While AFDB structures are cleaner and more idealized, PDB structures capture the intrinsic variability and physical realism essential for generalization in downstream tasks. To address this discrepancy, we introduce a Debiasing Structure AutoEncoder (DeSAE) that learns to reconstruct native-like conformations from intentionally corrupted backbone geometries. By training the model to recover plausible structural states, DeSAE implicitly captures a more robust and natural structural manifold. At inference, applying DeSAE to AFDB structures produces debiased structures that significantly improve inverse folding performance across multiple benchmarks. This work highlights the critical impact of subtle systematic biases in predicted structures and presents a principled framework for debiasing, significantly boosting the performance of structure-based learning tasks like inverse folding.

Depth3DLane: Monocular 3D Lane Detection via Depth Prior Distillation

Monocular 3D lane detection is challenging due to the difficulty in capturing depth information from single-camera images. A common strategy involves transforming front-view (FV) images into bird's-eye-view (BEV) space through inverse perspective mapping (IPM), facilitating lane detection using BEV features. However, IPM's flat-ground assumption and loss of contextual information lead to inaccuracies in reconstructing 3D information, especially height. In this paper, we introduce a BEV-based framework to address these limitations and improve 3D lane detection accuracy. Our approach incorporates a Hierarchical Depth-Aware Head that provides multi-scale depth features, mitigating the flat-ground assumption by enhancing spatial awareness across varying depths. Additionally, we leverage Depth Prior Distillation to transfer semantic depth knowledge from a teacher model, capturing richer structural and contextual information for complex lane structures. To further refine lane continuity and ensure smooth lane reconstruction, we introduce a Conditional Random Field module that enforces spatial coherence in lane predictions. Extensive experiments validate that our method achieves state-of-the-art performance in terms of z-axis error and outperforms other methods in the field in overall performance. The code is released at: https://anonymous.4open.science/r/Depth3DLane-DCDD.

MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and Quantization

Masked Image Modeling (MIM) with Vector Quantization (VQ) has achieved great success in both self-supervised pre-training and image generation. However, most existing methods struggle to address the trade-off in shared latent space for generation quality vs. representation learning and efficiency. To push the limits of this paradigm, we propose MergeVQ, which incorporates token merging techniques into VQ-based generative models to bridge the gap between image generation and visual representation learning in a unified architecture. During pre-training, MergeVQ decouples top-k semantics from latent space with the token merge module after self-attention blocks in the encoder for subsequent Look-up Free Quantization (LFQ) and global alignment and recovers their fine-grained details through cross-attention in the decoder for reconstruction. As for the second-stage generation, we introduce MergeAR, which performs KV Cache compression for efficient raster-order prediction. Extensive experiments on ImageNet verify that MergeVQ as an AR generative model achieves competitive performance in both visual representation learning and image generation tasks while maintaining favorable token efficiency and inference speed. The code and model will be available at https://apexgen-x.github.io/MergeVQ.

Specification Generation for Neural Networks in Systems

Specifications - precise mathematical representations of correct domain-specific behaviors - are crucial to guarantee the trustworthiness of computer systems. With the increasing development of neural networks as computer system components, specifications gain more importance as they can be used to regulate the behaviors of these black-box models. Traditionally, specifications are designed by domain experts based on their intuition of correct behavior. However, this is labor-intensive and hence not a scalable approach as computer system applications diversify. We hypothesize that the traditional (aka reference) algorithms that neural networks replace for higher performance can act as effective proxies for correct behaviors of the models, when available. This is because they have been used and tested for long enough to encode several aspects of the trustworthy/correct behaviors in the underlying domain. Driven by our hypothesis, we develop a novel automated framework, SpecTRA to generate specifications for neural networks using references. We formulate specification generation as an optimization problem and solve it with observations of reference behaviors. SpecTRA clusters similar observations into compact specifications. We present specifications generated by SpecTRA for neural networks in adaptive bit rate and congestion control algorithms. Our specifications show evidence of being correct and matching intuition. Moreover, we use our specifications to show several unknown vulnerabilities of the SOTA models for computer systems.

Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks

Modern AI agents, driven by advances in large foundation models, promise to enhance our productivity and transform our lives by augmenting our knowledge and capabilities. To achieve this vision, AI agents must effectively plan, perform multi-step reasoning and actions, respond to novel observations, and recover from errors, to successfully complete complex tasks across a wide range of scenarios. In this work, we introduce Magentic-One, a high-performing open-source agentic system for solving such tasks. Magentic-One uses a multi-agent architecture where a lead agent, the Orchestrator, plans, tracks progress, and re-plans to recover from errors. Throughout task execution, the Orchestrator directs other specialized agents to perform tasks as needed, such as operating a web browser, navigating local files, or writing and executing Python code. We show that Magentic-One achieves statistically competitive performance to the state-of-the-art on three diverse and challenging agentic benchmarks: GAIA, AssistantBench, and WebArena. Magentic-One achieves these results without modification to core agent capabilities or to how they collaborate, demonstrating progress towards generalist agentic systems. Moreover, Magentic-One's modular design allows agents to be added or removed from the team without additional prompt tuning or training, easing development and making it extensible to future scenarios. We provide an open-source implementation of Magentic-One, and we include AutoGenBench, a standalone tool for agentic evaluation. AutoGenBench provides built-in controls for repetition and isolation to run agentic benchmarks in a rigorous and contained manner -- which is important when agents' actions have side-effects. Magentic-One, AutoGenBench and detailed empirical performance evaluations of Magentic-One, including ablations and error analysis are available at https://aka.ms/magentic-one