Education
Unleashing Region Understanding in Intermediate Layers for MLLM-based Referring Expression Generation
On the other hand, complicated sentences could easily increase the probability of hallucinations.
Can Large Language Models Explore In-Context?
We focus on native performance of existing LLMs, without training interventions.
Label Delay in Online Continual Learning
We introduce a new continual learning framework with explicit modeling of the label delay between data and label streams over time steps.
AutoGuide: Automated Generation and Selection of Context-Aware Guidelines for Large Language Model Agents
Recent advances in large language models (LLMs) have empowered AI agents capable of performing various sequential decision-making tasks.
ALPINE: Unveiling The Planning Capability of Autoregressive Learning in Language Models Siwei Wang
Our mathematical characterization shows that Transformer architectures can execute path-finding by embedding the adjacency and reachability matrices within their weights. Furthermore, our theoretical analysis of gradient-based learning dynamics reveals that LLMs can learn both the adjacency and a limited form of the reachability matrices.
STL: Still Tricky Logic (for System Validation, Even When Showing Y our Work) Isabelle Hurley
Previous work showed that despite claims of interpretability, humans are unable to use formal specifications presented in a variety of ways to validate even simple robot behaviors.
Mixture of Experts Meets Prompt-Based Continual Learning Minh Le3 An Nguyen 2 Huy Nguyen 1 Trang Nguyen
However, current AI models often fail to retain this ability.
Superposed Decoding: Multiple Generations from a Single Autoregressive Inference Pass Ethan Shen Alan Fan Sarah Pratt Jae Sung Park Matthew Wallingford Sham Kakade Ari Holtzman Ranjay Krishna
Many applications today provide users with multiple auto-complete drafts as they type, including GitHub's code completion, Gmail's smart compose, and Apple's messaging auto-suggestions.
Inductive biases of multi-task learning and finetuning: multiple regimes of feature reuse
Neural networks are often trained on multiple tasks, either simultaneously (multi-task learning, MTL) or sequentially (pretraining and subsequent finetuning, PT+FT). In particular, it is common practice to pretrain neural networks on a large auxiliary task before finetuning on a downstream task with fewer samples. Despite the prevalence of this approach, the inductive biases that arise from learning multiple tasks are poorly characterized. In this work, we address this gap.