Qi Chen

This paper introduces a novel algorithm, CiSC, to efficiently and optimally synthesize circuit implementations of linear codes for hardware security, significantly outperforming existing state-of-the-art methods.

SkillAttack is a red-teaming framework that dynamically tests the exploitability of latent vulnerabilities in LLM agent skills using adversarial prompting, demonstrating that even benign skills pose significant security risks.

This paper introduces a novel denial-of-service attack targeting multi-round transaction simulation by exploiting inter-transaction dependencies within smart-contract state.

The paper compares various strong secrecy capacities for Arbitrarily Varying and General Arbitrarily Varying Wiretap Channels, establishing equivalences under specific conditions and bounding the capacity gap.

This paper enhances a genetic algorithm approach for solving the Shortest Vector Problem (SVP) in both integral and module lattices by incorporating domain-informed representation and crossover.

This paper enhances a genetic algorithm approach for solving the Shortest Vector Problem (SVP) in lattices by incorporating domain-informed representation, thereby extending its applicability to module lattices.

The paper proposes projectional decoding, a novel framework that integrates a partial graph model alongside text generation to ensure the semantic validity of LLM-generated software artifacts.

The paper addresses the challenge of multi-turn view planning for VLMs by proposing an iterative framework that uses self-exploration and view graph distillation, significantly improving planning performance over state-of-the-art models.

The paper introduces PithTrain, a compact, agent-native Mixture-of-Experts (MoE) training framework that significantly improves agent-task efficiency compared to existing production stacks.

The paper introduces Score-Guided Classification (SGC), a novel framework that uses an unsupervised anomaly score as a 'Pathological Prior' to guide EEG-based depression detection, overcoming the limitations of data augmentation in small-sample settings.

The paper introduces a new benchmark, E2V-Bench, to evaluate text-to-image models on generating pedagogically accurate visuals from arithmetic equations, finding that current models often fail due to structural and numerical errors.

The paper introduces BioConCal, a supervised scoring mechanism that evaluates biomedical NER candidates surfaced by multiple LLMs, significantly improving the quality of the candidate pool for human curators.

MOSS-Audio is a unified audio-language model designed for comprehensive understanding of speech, environmental sounds, and music, achieving strong performance across various audio-grounded tasks.

The paper introduces Tree-like Self-Play (TSP), a novel framework that treats secure code generation as a fine-grained decision process, significantly improving LLM security by forcing the model to self-correct localized vulnerabilities.

The paper builds a benchmark to evaluate the ability of multimodal large language models to extract accurate data tables from chart images, and proposes a human-centered approach to improve numerical accuracy.

MagicSelector is a framework for tool retrieval in agents using counterfactual task decomposition, progressive reranking, and dynamic Top-K.

This paper benchmarks zero-shot synthesis of parent-selection operators across eight large language models and finds that Claude Sonnet~4.6 and Gemini~3.1 Pro perform strongly, with the best operator surpassing automatic baselines.