Question Answering

Transformers excel at graph reasoning, with logarithmic depth proving necessary and sufficient for parallelizable tasks; single-layer transformers solve retrieval tasks efficiently.

TableRAG, a novel Retrieval-Augmented Generation framework, achieves state-of-the-art performance in large-scale table understanding by efficiently integrating schema and cell retrieval with language …

SpikedAttention: Training-free transformer-to-SNN conversion achieving state-of-the-art accuracy and 42% energy reduction!

RG-SAN achieves state-of-the-art 3D referring expression segmentation by leveraging spatial awareness and rule-guided weak supervision, significantly improving accuracy and handling of ambiguous descr…

RankRAG: One LLM, dual-purpose instruction-tuning for superior RAG!

Plan-on-Graph (PoG) revolutionizes KG-augmented LLMs with a self-correcting adaptive planning paradigm, enabling more efficient and accurate reasoning over knowledge graphs by dynamically adjusting ex…

MEDIQ benchmark revolutionizes LLM evaluation by shifting from static to interactive clinical reasoning, revealing LLMs’ struggles with proactive information-seeking and highlighting the importance of…

MDAgents: An adaptive multi-agent LLM framework boosts medical decision-making accuracy by dynamically adjusting collaboration structures based on task complexity.

New dataset MUSIC-AVQA-R and a multi-faceted cycle collaborative debiasing strategy significantly improve audio-visual question answering robustness.

LIVE, a novel learnable in-context vector, significantly improves visual question answering by reducing computational costs and enhancing accuracy compared to traditional ICL methods.

KnowGPT: A novel framework boosts Large Language Model accuracy by intelligently integrating knowledge graphs, significantly reducing factual errors and achieving near-human performance on benchmark d…

IQA-EVAL: An automatic evaluation framework uses LLMs to simulate human-AI interactions and evaluate interactive question answering, achieving high correlation with human judgments.

G-Retriever: a novel RAG approach enables conversational interaction with textual graphs, improving graph understanding and question answering efficiency while mitigating hallucination.

DeAR: A novel framework lets LLMs solve complex problems with human-like iterative reasoning.

Coke: A cost-efficient KBQA strategy using LLMs and KGMs, maximizing accuracy while minimizing GPT-4 fees by up to 20.89%

Conformal Alignment certifies trustworthy foundation model outputs by guaranteeing a user-specified fraction meet alignment criteria, regardless of the model or data.

ChatQA, a new suite of models, outperforms GPT-4 in conversational QA and RAG by using a two-stage instruction tuning method and a cost-effective dense retriever.

AMOR: Adaptable Modular knowledge agent using LLMs, excels with FSM-based reasoning and process feedback, enabling human supervision and domain adaptation.

AGILE, a novel reinforcement learning framework, significantly enhances LLM agents’ performance on complex conversational tasks by integrating memory, tools, expert interactions, and reflection, outpe…

KG-ICL, a novel prompt-based knowledge graph foundation model, achieves universal in-context reasoning by leveraging in-context learning and a unified tokenizer, outperforming various baselines on 43 …