Paper-Conference

Conversing with the Past: Immersive Access to Cultural Digital Archives with LLMs and Spatial Interfaces
An IUI Companion 2026 paper on LLM-powered spatial access to cultural archives.
Cite DOI ACM DL ACM DOI
Conversing with the Past: Immersive Access to Cultural Digital Archives with LLMs and Spatial Interfaces
MuSLR: Multimodal Symbolic Logical Reasoning
A NeurIPS 2025 benchmark and method for multimodal symbolic logical reasoning.
PDF Cite NeurIPS Project
MuSLR: Multimodal Symbolic Logical Reasoning
SkyLadder: Better and Faster Pretraining via Context Window Scheduling
A NeurIPS 2025 method for faster and stronger pretraining through context window scheduling.
PDF Cite Code NeurIPS
SkyLadder: Better and Faster Pretraining via Context Window Scheduling
KAHAN: Knowledge-Augmented Hierarchical Analysis and Narration for Financial Data Narration
A knowledge-augmented hierarchical framework that extracts insights from tabular data at entity, pairwise, group, and system levels for financial data narration.
PDF Cite Code Project DOI URL
Discursive Circuits: How Do Language Models Understand Discourse Relations?
We introduce a new task, Completion under Discourse Relation (CuDR), to reveal the underlying transformer circuits responsible for discourse understanding.
PDF Cite Code Project Slides
Discursive Circuits: How Do Language Models Understand Discourse Relations?
Beyond In-Context Learning: Aligning Long-form Generation of Large Language Models via Task-Inherent Attribute Guidelines
LongGuide generates task-inherent metric and output-constraint guidelines to improve long-form LLM generation beyond demonstrations alone.
PDF Cite
Beyond In-Context Learning: Aligning Long-form Generation of Large Language Models via Task-Inherent Attribute Guidelines
What Makes a Good Natural Language Prompt?
A meta-analysis of prompting research that organizes natural-language prompt quality into 21 properties across six dimensions.
PDF Cite
What Makes a Good Natural Language Prompt?
LLMs Are Biased Towards Output Formats! Systematically Evaluating and Mitigating Output Format Bias of LLMs
A systematic evaluation of output-format bias in LLMs, with prompting and fine-tuning strategies for reducing format sensitivity.
PDF Cite
LLMs Are Biased Towards Output Formats! Systematically Evaluating and Mitigating Output Format Bias of LLMs
Aligning Large Language Models with Human Opinions through Persona Selection and Value-Belief-Norm Reasoning
Chain-of-Opinion selects and reasons over explicit and implicit personae to improve LLM opinion prediction and alignment.
Cite URL
Aligning Large Language Models with Human Opinions through Persona Selection and Value-Belief-Norm Reasoning
DnA-Eval: Enhancing Large Language Model Evaluation through Decomposition and Aggregation
DnA-Eval decomposes and aggregates LLM-as-judge evaluation to improve interpretability and agreement with human preferences.
Cite URL
DnA-Eval: Enhancing Large Language Model Evaluation through Decomposition and Aggregation