About
Research & Development
Evaluating Hallucinations in Vision-Language Models for Chart Interpretation
TLDR: CHARTHAL is a new benchmark that evaluates how Large Vision-Language Models (LVLMs) hallucinate when understanding charts. It uses a fine-grained taxonomy of question...
Research & Development
MapCoder-Lite: Empowering Small Language Models for Advanced Code Generation
TLDR: MapCoder-Lite is a novel framework that enables a small 7-billion-parameter (7B) language model to perform high-quality multi-agent code generation for competitive programming. It...
Research & Development
Instruction Tuning Large Language Models for Better Coreference Resolution Across Languages
TLDR: The paper introduces CorefInst, the first multilingual coreference resolution (CR) method that uses instruction-tuned, decoder-only Large Language Models (LLMs) to handle both explicit...
Research & Development
LLMs Adapt to Non-Text Data Through In-Context Representation Learning
TLDR: A new method called In-Context Representation Learning (ICRL) allows Large Language Models (LLMs) to process non-text data (like molecular structures, images, or audio)...
Research & Development
Enhancing YOLO’s Reliability for Underwater Object Detection
TLDR: This study empirically evaluates the robustness of YOLOv8-YOLOv12 models for underwater object detection across various simulated distortions (noise, blur, color shifts). It finds...
Research & Development
MSCoRe: A New Benchmark for Evaluating Multi-Stage Reasoning in LLM Agents
TLDR: MSCoRe is a novel benchmark featuring 126,696 QA instances across automotive, pharmaceutical, electronics, and energy sectors, designed to assess LLM agents' multi-stage collaborative...