Posters

LaSe-E2V: Language-guided semantic-aware event-to-video reconstruction uses text descriptions to improve video quality and consistency.

LaSCal, a novel label-free calibration method, ensures reliable model predictions under label shift by using a consistent calibration error estimator, achieving effective and robust unsupervised calib…

Large stepsize GD on non-homogeneous neural networks shows monotonic risk reduction after an initial oscillating phase, demonstrating implicit bias and optimization gains.

Large Spatial Model (LSM) achieves real-time semantic 3D reconstruction from just two unposed images, unifying multiple 3D vision tasks in a single feed-forward pass.

TABULA-8B, a novel language model for tabular prediction, achieves state-of-the-art zero-shot and few-shot performance across various benchmarks, exceeding existing methods by 5-15 percentage points.

Large Pre-trained Time-series Models (LPTM) achieves superior forecasting and time-series classification results using a novel adaptive segmentation method, requiring up to 40% less data and 50% less …

LLM-DA dynamically adapts LLM-generated rules for accurate, interpretable temporal knowledge graph reasoning, significantly improving accuracy without fine-tuning.

LLMs conquer StarCraft II: A new benchmark and Chain of Summarization method enable real-time strategic gameplay evaluation, showcasing impressive LLM strategic abilities.

Teach LLMs uncertainty for reliable high-stakes predictions: Fine-tuning with graded examples significantly improves LLM’s uncertainty calibration and generalizes well.

LLM agents effectively generate realistic personal mobility patterns using semantically rich data.

New conformal inference methods enhance LLM validity by providing adaptive validity guarantees and improving the quality of LLM outputs, addressing prior methods’ limitations.

ECO prompts enable efficient LLM unlearning by corrupting prompts flagged for forgetting, achieving promising results across various LLMs and tasks with minimal side effects.

This paper presents a novel method for large language model (LLM) unlearning, enabling LLMs to ‘forget’ undesirable behaviors by using only negative examples. This computationally efficient approach o…

InteractTraj: Generating realistic, interactive traffic trajectories from natural language!

Large language models (LLMs) achieve lossless gradient compression, surpassing existing methods by up to 17.2%, thereby advancing distributed learning efficiency.

Language models struggle with hierarchical information. This work introduces Hierarchy Transformer Encoders (HITs), a novel method to retrain transformer encoders using hyperbolic geometry and special…

LangGround: MARL agents learn human-interpretable communication via LLM-grounded training, enabling effective human-agent collaboration.

Lambda: A novel framework tackles entity alignment challenges with unlabeled dangling entities using GNN-based encoding, spectral contrastive learning, and an iterative PU learning algorithm, achievin…

LAM3D: A novel framework uses point cloud data to boost single-image 3D mesh reconstruction accuracy, achieving state-of-the-art results in just 6 seconds.

LaKD: a novel length-agnostic knowledge distillation framework enables accurate trajectory prediction regardless of observation length, overcoming limitations of existing methods.