arXiv subjects

Yujun Sun

Publications and source records attributed to Yujun Sun.

4 recordsLinked to original sources

arXiv2026

oxo-call: Documentation-grounded Skill Augmentation for Accurate Bioinformatics Command-line Generation with Large Language Models

Command-line bioinformatics tools remain essential for genomic analysis, yet their diversity in syntax and parameterization presents a persistent barrier to productive research. We present oxo-call, a Rust-based command-line assistant that translates natural-language task descriptions into accurate tool invocations through two complementary strategies: documentation-first grounding, which provides the large language model (LLM) with the complete, version-specific help text of each target tool, and curated skill augmentation, which primes the model with domain-expert concepts, common pitfalls, and worked examples. oxo-call (v0.10) ships >150 built-in skills covering 44 analytical categories, from variant calling and genome assembly to single-cell transcriptomics, compiled into a single, statically linked binary. Every generated command is logged with provenance metadata to support reproducible research. oxo-call also provides a DAG-based workflow engine, extensibility through user-defined and community skills via the Model Context Protocol, and support for local LLM inference to address data-privacy requirements. oxo-call is freely available for academic use at https://traitome.github.io/oxo-call/.

q-bio.GN↗

arXiv2026

ReLoop: Structured Modeling and Behavioral Verification for Reliable LLM-Based Optimization

Large language models (LLMs) can translate natural language into optimization code, but silent failures pose a critical risk: code that executes and returns solver-feasible solutions may encode semantically incorrect formulations---a feasibility--correctness gap reaching 90 percentage points on compositional problems. We introduce ReLoop, which addresses this gap through two complementary mechanisms. Structured generation decomposes code production into a four-stage reasoning chain (understand, formalize, synthesize, verify), preventing formulation errors at their source. Behavioral verification detects errors that survive generation by testing whether the formulation responds correctly to solver-based parameter perturbation---an external semantic signal that bypasses LLM self-review and requires no ground truth. The two mechanisms are complementary by error structure: structured generation drives the largest gains on compositional problems (+8.5pp accuracy on RetailOpt-190 with Claude Opus 4.6), while behavioral verification dominates on localized defects +4.4pp on MAMO-ComplexLP, its largest contribution across benchmarks). Combined with diagnostic execution recovery, ReLoop reaches 100% executable code on Claude Opus 4.6 and consistently improves accuracy on chat-tuned foundation models across three benchmarks; we further identify a known limitation of narrowly-tuned SFT models, whose learned output formats are brittle to chain-of-thought prompts---an interaction we document and analyze. We release RetailOpt-190, 190 compositional retail optimization scenarios targeting the multi-constraint interactions where LLMs most frequently fail.

cs.SE↗

arXiv2026

TwistNet-2D: Learning Second-Order Channel Interactions via Spiral Twisting for Texture Recognition

Second-order feature statistics are central to texture recognition, yet existing mechanisms exhibit a structural tension: bilinear pooling and Gram matrices capture global channel correlations but discard spatial structure, whereas self-attention models capture cross-position relations through weighted sums rather than explicit pairwise products. We propose TwistNet-2D, a lightweight module that computes local pairwise channel products under directional spatial displacement, jointly encoding where features co-occur and how they interact. The core component, Spiral-Twisted Channel Interaction (STCI), shifts one feature map along a prescribed direction before L2-normalized channel multiplication, capturing cross-position co-occurrence patterns that characterize structured and periodic textures. Four directional heads are aggregated through content-adaptive channel reweighting, and the result is injected via a sigmoid-gated residual path with near-zero initialization. TwistNet-2D adds only approximately 3.5% parameters and approximately 2% FLOPs over ResNet-18. To isolate the contribution of architectural inductive bias from that of transfer learning, all models in this study are trained from scratch without ImageNet pretraining. Under this protocol, TwistNet-2D consistently surpasses parameter-matched baselines and substantially larger ConvNeXt and Swin Transformer backbones across four texture and fine-grained recognition benchmarks, while the multi-head structure produces interpretable, orientation-selective representations that align with classical texture analysis.

cs.CV↗

arXiv2025

CIP: A Plug-and-Play Causal Prompting Framework for Mitigating Hallucinations under Long-Context Noise

Large language models often hallucinate when processing long and noisy retrieval contexts because they rely on spurious correlations rather than genuine causal relationships. We propose CIP, a lightweight and plug-and-play causal prompting framework that mitigates hallucinations at the input stage. CIP constructs a causal relation sequence among entities, actions, and events and injects it into the prompt to guide reasoning toward causally relevant evidence. Through causal intervention and counterfactual reasoning, CIP suppresses non causal reasoning paths, improving factual grounding and interpretability. Experiments across seven mainstream language models, including GPT-4o, Gemini 2.0 Flash, and Llama 3.1, show that CIP consistently enhances reasoning quality and reliability, achieving 2.6 points improvement in Attributable Rate, 0.38 improvement in Causal Consistency Score, and a fourfold increase in effective information density. API level profiling further shows that CIP accelerates contextual understanding and reduces end to end response latency by up to 55.1 percent. These results suggest that causal reasoning may serve as a promising paradigm for improving the explainability, stability, and efficiency of large language models.

cs.CL↗