arXiv subjects

Yang Yang

Publications and source records attributed to Yang Yang.

At least 433 records · Page 24Linked to original sources

arXiv2025

eSapiens: A Platform for Secure and Auditable Retrieval-Augmented Generation

We present eSapiens, an AI-as-a-Service (AIaaS) platform engineered around a business-oriented trifecta: proprietary data, operational workflows, and any major agnostic Large Language Model (LLM). eSapiens gives businesses full control over their AI assets, keeping everything in-house for AI knowledge retention and data security. eSapiens AI Agents (Sapiens) empower your team by providing valuable insights and automating repetitive tasks, enabling them to focus on high-impact work and drive better business outcomes. The system integrates structured document ingestion, hybrid vector retrieval, and no-code orchestration via LangChain, and supports top LLMs including OpenAI, Claude, Gemini, and DeepSeek. A key component is the THOR Agent, which handles structured SQL-style queries and generates actionable insights over enterprise databases. To evaluate the system, we conduct two experiments. First, a retrieval benchmark on legal corpora reveals that a chunk size of 512 tokens yields the highest retrieval precision (Top-3 accuracy: 91.3%). Second, a generation quality test using TRACe metrics across five LLMs shows that eSapiens delivers more context-consistent outputs with up to a 23% improvement in factual alignment. These results demonstrate the effectiveness of eSapiens in enabling trustworthy, auditable AI workflows for high-stakes domains like legal and finance.

cs.AI↗

arXiv2025

eSapiens's DEREK Module: Deep Extraction & Reasoning Engine for Knowledge with LLMs

We present the DEREK (Deep Extraction & Reasoning Engine for Knowledge) Module, a secure and scalable Retrieval-Augmented Generation pipeline designed specifically for enterprise document question answering. Designed and implemented by eSapiens, the system ingests heterogeneous content (PDF, Office, web), splits it into 1,000-token overlapping chunks, and indexes them in a hybrid HNSW+BM25 store. User queries are refined by GPT-4o, retrieved via combined vector+BM25 search, reranked with Cohere, and answered by an LLM using CO-STAR prompt engineering. A LangGraph verifier enforces citation overlap, regenerating answers until every claim is grounded. On four LegalBench subsets, 1000-token chunks improve Recall@50 by approximately 1 pp and hybrid+rerank boosts Precision@10 by approximately 7 pp; the verifier raises TRACe Utilization above 0.50 and limits unsupported statements to less than 3%. All components run in containers, enforce end-to-end TLS 1.3 and AES-256. These results demonstrate that the DEREK module delivers accurate, traceable, and production-ready document QA with minimal operational overhead. The module is designed to meet enterprise demands for secure, auditable, and context-faithful retrieval, providing a reliable baseline for high-stakes domains such as legal and finance.

cs.CL↗

arXiv2025

MedSegFactory: Text-Guided Generation of Medical Image-Mask Pairs

This paper presents MedSegFactory, a versatile medical synthesis framework that generates high-quality paired medical images and segmentation masks across modalities and tasks. It aims to serve as an unlimited data repository, supplying image-mask pairs to enhance existing segmentation tools. The core of MedSegFactory is a dual-stream diffusion model, where one stream synthesizes medical images and the other generates corresponding segmentation masks. To ensure precise alignment between image-mask pairs, we introduce Joint Cross-Attention (JCA), enabling a collaborative denoising paradigm by dynamic cross-conditioning between streams. This bidirectional interaction allows both representations to guide each other's generation, enhancing consistency between generated pairs. MedSegFactory unlocks on-demand generation of paired medical images and segmentation masks through user-defined prompts that specify the target labels, imaging modalities, anatomical regions, and pathological conditions, facilitating scalable and high-quality data generation. This new paradigm of medical image synthesis enables seamless integration into diverse medical imaging workflows, enhancing both efficiency and accuracy. Extensive experiments show that MedSegFactory generates data of superior quality and usability, achieving competitive or state-of-the-art performance in 2D and 3D segmentation tasks while addressing data scarcity and regulatory constraints.

cs.CV↗

arXiv2025

Pushing the Limits of Safety: A Technical Report on the ATLAS Challenge 2025

Multimodal Large Language Models (MLLMs) have enabled transformative advancements across diverse applications but remain susceptible to safety threats, especially jailbreak attacks that induce harmful outputs. To systematically evaluate and improve their safety, we organized the Adversarial Testing & Large-model Alignment Safety Grand Challenge (ATLAS) 2025}. This technical report presents findings from the competition, which involved 86 teams testing MLLM vulnerabilities via adversarial image-text attacks in two phases: white-box and black-box evaluations. The competition results highlight ongoing challenges in securing MLLMs and provide valuable guidance for developing stronger defense mechanisms. The challenge establishes new benchmarks for MLLM safety evaluation and lays groundwork for advancing safer multimodal AI systems. The code and data for this challenge are openly available at https://github.com/NY1024/ATLAS_Challenge_2025.

cs.CR↗

arXiv2025

Observation of superconductivity-induced leading-edge gap in Sr-doped $\mathrm{La}_{3}\mathrm{Ni}_{2}\mathrm{O}_{7}$ thin films

The discovery of high-temperature superconductivity in pressurized bulk $\mathrm{La}_{3}\mathrm{Ni}_{2}\mathrm{O}_{7}$ has ignited significant interest in nickelate superconductors. Unlike cuprates, where superconductivity predominantly originates from the $\mathrm{3}d_{x^2-y^2}$ orbital, nickelates exhibit additional complexities involving contributions from the $\mathrm{3}d_{z^2}$ orbital, prompting fundamental questions about their pairing mechanisms. Despite recent progress in stabilizing superconductivity in $\mathrm{La}_{3}\mathrm{Ni}_{2}\mathrm{O}_{7}$ thin films at ambient pressure, direct spectroscopic evidence of the superconducting gap opening remains elusive. Here, we present an in-situ angle-resolved photoemission spectroscopy study of Sr-doped superconducting $\mathrm{La}_{3}\mathrm{Ni}_{2}\mathrm{O}_{7}$ thin films. Fermi surface mapping reveals Ni-$\mathrm{3}d_{x^2-y^2}$-derived $α$ and $β$ pockets, with orbital fillings of 0.11$\pm$0.02 electrons and 0.66$\pm$0.03 holes per Ni, respectively, resulting in a total of 0.45$\pm$0.04 electrons for each Ni. These bands exhibit moderate electron correlations, characterized by a band renormalization factor of 3-4. Notably, both $α$ and $β$ bands exhibit leading-edge shifts across the superconducting transition, with gap magnitude of ~1-2 meV at Fermi momenta along the Brillouin zone diagonal and slightly away from the zone diagonal, deviating from the conventional $d_{x^2-y^2}$-wave gap structure. Additionally, the Ni-$\mathrm{3}d_{z^2}$-derived $γ$ band lies ~75 meV below the Fermi level, indicating a $\mathrm{3}d_{x^2-y^2}$-dominated fermiology in this compound.

cond-mat.supr-con↗

arXiv2025

Learning Segmentation from Radiology Reports

Tumor segmentation in CT scans is key for diagnosis, surgery, and prognosis, yet segmentation masks are scarce because their creation requires time and expertise. Public abdominal CT datasets have from dozens to a couple thousand tumor masks, but hospitals have hundreds of thousands of tumor CTs with radiology reports. Thus, leveraging reports to improve segmentation is key for scaling. In this paper, we propose a report-supervision loss (R-Super) that converts radiology reports into voxel-wise supervision for tumor segmentation AI. We created a dataset with 6,718 CT-Report pairs (from the UCSF Hospital), and merged it with public CT-Mask datasets (from AbdomenAtlas 2.0). We used our R-Super to train with these masks and reports, and strongly improved tumor segmentation in internal and external validation--F1 Score increased by up to 16% with respect to training with masks only. By leveraging readily available radiology reports to supplement scarce segmentation masks, R-Super strongly improves AI performance both when very few training masks are available (e.g., 50), and when many masks were available (e.g., 1.7K). Project: https://github.com/MrGiovanni/R-Super

eess.IV↗

arXiv2025

Uncertainty-aware Reward Design Process

Designing effective reward functions is a cornerstone of reinforcement learning (RL), yet it remains a challenging process due to the inefficiencies and inconsistencies inherent in conventional reward engineering methodologies. Recent advances have explored leveraging large language models (LLMs) to automate reward function design. However, their suboptimal performance in numerical optimization often yields unsatisfactory reward quality, while the evolutionary search paradigm demonstrates inefficient utilization of simulation resources, resulting in prohibitively lengthy design cycles with disproportionate computational overhead. To address these challenges, we propose the Uncertainty-aware Reward Design Process (URDP), a novel framework that integrates large language models to streamline reward function design and evaluation in RL environments. URDP quantifies candidate reward function uncertainty based on self-consistency analysis, enabling simulation-free identification of ineffective reward components while discovering novel reward components. Furthermore, we introduce uncertainty-aware Bayesian optimization (UABO), which incorporates uncertainty estimation to significantly enhance hyperparameter configuration efficiency. Finally, we construct a bi-level optimization architecture by decoupling the reward component optimization and the hyperparameter tuning. URDP orchestrates synergistic collaboration between the reward logic reasoning of the LLMs and the numerical optimization strengths of the Bayesian Optimization. We conduct a comprehensive evaluation of URDP across 35 diverse tasks spanning three benchmark environments. Our experimental results demonstrate that URDP not only generates higher-quality reward functions but also achieves significant improvements in the efficiency of automated reward design compared to existing approaches.

cs.LG↗

arXiv2025

Thermal Quench Dynamics of Visons in Gapless Kitaev Spin Liquid

The relaxation dynamics of the Kitaev honeycomb model under a thermal quench is dominated by the quasi-stochastic diffusion and pair annihilation of visions, which are gapped flux excitations of an emergent $\mathbb{Z}_2$ gauge field of the Kitaev spin liquid. Both the diffusion energy barrier as well as the effective interactions between visons are mediated by the Majorana fermions which are fractionalized quasiparticles of the spin liquid. Through extensive kinetic Monte Carlo simulations, we show that the interplay between the thermal diffusion and nonlocal multi-vision interactions leads to a variety of temperature-dependent dynamical behaviors ranging from diffusion-limited and terminal-velocity-limited annihilation to dynamical arresting and freezing. Notably, we show that the freezing phenomenon is intimately related to the formation of metastable $\sqrt{3}\times\sqrt{3}$ vison crystals and a hidden coarsening of super-clusters associated with a broken $\mathbb{Z}_3$ symmetry.

cond-mat.str-el↗

arXiv2025

PanTS: The Pancreatic Tumor Segmentation Dataset

PanTS is a large-scale, multi-institutional dataset curated to advance research in pancreatic CT analysis. It contains 36,390 CT scans from 145 medical centers, with expert-validated, voxel-wise annotations of over 993,000 anatomical structures, covering pancreatic tumors, pancreas head, body, and tail, and 24 surrounding anatomical structures such as vascular/skeletal structures and abdominal/thoracic organs. Each scan includes metadata such as patient age, sex, diagnosis, contrast phase, in-plane spacing, slice thickness, etc. AI models trained on PanTS achieve significantly better performance in pancreatic tumor detection, localization, and segmentation compared to those trained on existing public datasets. Our analysis indicates that these gains are directly attributable to the 16x larger-scale tumor annotations and indirectly supported by the 24 additional surrounding anatomical structures. As the largest and most comprehensive resource of its kind, PanTS offers a new benchmark for developing and evaluating AI models in pancreatic CT analysis.

eess.IV↗

arXiv2025

Towards Competitive Search Relevance For Inference-Free Learned Sparse Retrievers

Learned sparse retrieval, which can efficiently perform retrieval through mature inverted-index engines, has garnered growing attention in recent years. Particularly, the inference-free sparse retrievers are attractive as they eliminate online model inference in the retrieval phase thereby avoids huge computational cost, offering reasonable throughput and latency. However, even the state-of-the-art (SOTA) inference-free sparse models lag far behind in terms of search relevance when compared to both sparse and dense siamese models. Towards competitive search relevance for inference-free sparse retrievers, we argue that they deserve dedicated training methods other than using same ones with siamese encoders. In this paper, we propose two different approaches for performance improvement. First, we propose an IDF-aware penalty for the matching function that suppresses the contribution of low-IDF tokens and increases the model's focus on informative terms. Moreover, we propose a heterogeneous ensemble knowledge distillation framework that combines siamese dense and sparse retrievers to generate supervisory signals during the pre-training phase. The ensemble framework of dense and sparse retriever capitalizes on their strengths respectively, providing a strong upper bound for knowledge distillation. To concur the diverse feedback from heterogeneous supervisors, we normalize and then aggregate the outputs of the teacher models to eliminate score scale differences. On the BEIR benchmark, our model outperforms existing SOTA inference-free sparse model by \textbf{3.3 NDCG@10 score}. It exhibits search relevance comparable to siamese sparse retrievers and client-side latency only \textbf{1.1x that of BM25}.

cs.IR↗

arXiv2025

Generalizing vision-language models to novel domains: A comprehensive survey

Recently, vision-language pretraining has emerged as a transformative technique that integrates the strengths of both visual and textual modalities, resulting in powerful vision-language models (VLMs). Leveraging web-scale pretraining data, these models exhibit strong zero-shot capabilities. However, their performance often deteriorates when confronted with domain-specific or specialized generalization tasks. To address this, a growing body of research focuses on transferring or generalizing the rich knowledge embedded in VLMs to various downstream applications. This survey aims to comprehensively summarize the generalization settings, methodologies, benchmarking and results in VLM literatures. Delving into the typical VLM structures, current literatures are categorized into prompt-based, parameter-based and feature-based methods according to the transferred modules. The differences and characteristics in each category are furthered summarized and discussed by revisiting the typical transfer learning (TL) settings, providing novel interpretations for TL in the era of VLMs. Popular benchmarks for VLM generalization are further introduced with thorough performance comparisons among the reviewed methods. Following the advances in large-scale generalizable pretraining, this survey also discusses the relations and differences between VLMs and up-to-date multimodal large language models (MLLM), e.g., DeepSeek-VL. By systematically reviewing the surging literatures in vision-language research from a novel and practical generalization prospective, this survey contributes to a clear landscape of current and future multimodal researches.

cs.CV↗

arXiv2025

DiffSoundStream: Efficient Speech Tokenization via Diffusion Decoding

Token-based language modeling is a prominent approach for speech generation, where tokens are obtained by quantizing features from self-supervised learning (SSL) models and extracting codes from neural speech codecs, generally referred to as semantic tokens and acoustic tokens. These tokens are often modeled autoregressively, with the inference speed being constrained by the token rate. In this work, we propose DiffSoundStream, a solution that improves the efficiency of speech tokenization in non-streaming scenarios through two techniques: (1) conditioning the neural codec on semantic tokens to minimize redundancy between semantic and acoustic tokens, and (2) leveraging latent diffusion models to synthesize high-quality waveforms from semantic and coarse-level acoustic tokens. Experiments show that at 50 tokens per second, DiffSoundStream achieves speech quality on par with a standard SoundStream model operating at twice the token rate. Additionally, we achieve step-size distillation using just four diffusion sampling steps with only a minor quality loss.

eess.AS↗

arXiv2025

eSapiens: A Real-World NLP Framework for Multimodal Document Understanding and Enterprise Knowledge Processing

We introduce eSapiens, a unified question-answering system designed for enterprise settings, which bridges structured databases and unstructured textual corpora via a dual-module architecture. The system combines a Text-to-SQL planner with a hybrid Retrieval-Augmented Generation (RAG) pipeline, enabling natural language access to both relational data and free-form documents. To enhance answer faithfulness, the RAG module integrates dense and sparse retrieval, commercial reranking, and a citation verification loop that ensures grounding consistency. We evaluate eSapiens on the RAGTruth benchmark across five leading large language models (LLMs), analyzing performance across key dimensions such as completeness, hallucination, and context utilization. Results demonstrate that eSapiens outperforms a FAISS baseline in contextual relevance and generation quality, with optional strict-grounding controls for high-stakes scenarios. This work provides a deployable framework for robust, citation-aware question answering in real-world enterprise applications.

cs.IR↗

arXiv2025

Feynman-Kac Formula for Time-Dependent Nonlinear Schrödinger Equations with Applications in Numerical Approximations

In this paper, we present a novel Feynman-Kac formula and investigate learning-based methods for approximating general nonlinear time-dependent Schrödinger equations which may be high-dimensional. Our formulation integrates both the Fisk-Stratonovich and Itô integrals within the framework of backward stochastic differential equations (BSDEs). Utilizing this Feynman-Kac representation, we propose learning-based approaches for numerical approximations. To demonstrate the accuracy and effectiveness of the proposed method, we conduct numerical experiments in both low- and high-dimensional settings, complemented by a convergence analysis. These results address the open problem concerning deep-BSDE methods for numerical approximations of high-dimensional time-dependent nonlinear Schrödinger equations (cf. [Proc. Natl. Acad. Sci. 15 (2018), pp. 8505-8510] and [Frontiers Sci. Awards Math. (2024), pp. 1-14] by Han, Jentzen, and E).

math.AP↗

arXiv2025

SimuPanel: A Novel Immersive Multi-Agent System to Simulate Interactive Expert Panel Discussion

Panel discussion allows the audience to learn different perspectives through interactive discussions among experts moderated by a host and a Q&A session with the audience. Despite its benefits, panel discussion in the real world is inaccessible to many who do not have the privilege to participate due to geographical, financial, and time constraints. We present SimuPanel, which simulates panel discussions among academic experts through LLM-based multi-agent interaction. It enables users to define topics of interest for the panel, observe the expert discussion, engage in Q&A, and take notes. SimuPanel employs a host-expert architecture where each panel member is simulated by an agent with specialized expertise, and the panel is visualized in an immersive 3D environment to enhance engagement. Traditional dialogue generation struggles to capture the depth and interactivity of real-world panel discussions. To address this limitation, we propose a novel multi-agent interaction framework that simulates authentic panel dynamics by modeling reasoning strategies and personas of experts grounded in multimedia sources. This framework enables agents to dynamically recall and contribute to the discussion based on past experiences from diverse perspectives. Our technical evaluation and the user study with university students show that SimuPanel was able to simulate more in-depth discussions and engage participants to interact with and reflect on the discussions. As a first step in this direction, we offer design implications for future avenues to improve and harness the power of panel discussion for multimedia learning.

cs.HC↗

arXiv2025

Finance Language Model Evaluation (FLaME)

Language Models (LMs) have demonstrated impressive capabilities with core Natural Language Processing (NLP) tasks. The effectiveness of LMs for highly specialized knowledge-intensive tasks in finance remains difficult to assess due to major gaps in the methodologies of existing evaluation frameworks, which have caused an erroneous belief in a far lower bound of LMs' performance on common Finance NLP (FinNLP) tasks. To demonstrate the potential of LMs for these FinNLP tasks, we present the first holistic benchmarking suite for Financial Language Model Evaluation (FLaME). We are the first research paper to comprehensively study LMs against 'reasoning-reinforced' LMs, with an empirical study of 23 foundation LMs over 20 core NLP tasks in finance. We open-source our framework software along with all data and results.

cs.CL↗

arXiv2025

Lightweight Task-Oriented Semantic Communication Empowered by Large-Scale AI Models

Recent studies have focused on leveraging large-scale artificial intelligence (LAI) models to improve semantic representation and compression capabilities. However, the substantial computational demands of LAI models pose significant challenges for real-time communication scenarios. To address this, this paper proposes utilizing knowledge distillation (KD) techniques to extract and condense knowledge from LAI models, effectively reducing model complexity and computation latency. Nevertheless, the inherent complexity of LAI models leads to prolonged inference times during distillation, while their lack of channel awareness compromises the distillation performance. These limitations make standard KD methods unsuitable for task-oriented semantic communication scenarios. To address these issues, we propose a fast distillation method featuring a pre-stored compression mechanism that eliminates the need for repetitive inference, significantly improving efficiency. Furthermore, a channel adaptive module is incorporated to dynamically adjust the transmitted semantic information based on varying channel conditions, enhancing communication reliability and adaptability. In addition, an information bottleneck-based loss function is derived to guide the fast distillation process. Simulation results verify that the proposed scheme outperform baselines in term of task accuracy, model size, computation latency, and training data requirements.

cs.LG↗

arXiv2025

New Dataset and Methods for Fine-Grained Compositional Referring Expression Comprehension via Specialist-MLLM Collaboration

Referring Expression Comprehension (REC) is a foundational cross-modal task that evaluates the interplay of language understanding, image comprehension, and language-to-image grounding. It serves as an essential testing ground for Multimodal Large Language Models (MLLMs). To advance this field, we introduced a new REC dataset in our previous conference paper, characterized by two key features. First, it is designed with controllable difficulty levels, requiring multi-level fine-grained reasoning across object categories, attributes, and multi-hop relationships. Second, it incorporates negative text and images generated through fine-grained editing and augmentation, explicitly testing a model's ability to reject scenarios where the target object is absent, an often overlooked yet critical challenge in existing datasets. In this extended work, we propose two new methods to tackle the challenges of fine-grained REC by combining the strengths of Specialist Models and MLLMs. The first method adaptively assigns simple cases to faster, lightweight models and reserves complex ones for powerful MLLMs, balancing accuracy and efficiency. The second method lets a specialist generate a set of possible object regions, and the MLLM selects the most plausible one using its reasoning ability. These collaborative strategies lead to significant improvements on our dataset and other challenging benchmarks. Our results show that combining specialized and general-purpose models offers a practical path toward solving complex real-world vision-language tasks. Our dataset and code are available at https://github.com/sleepyshep/FineCops-Ref.

cs.CV↗