9 accepted papers
Visual reinforcement learning (RL) suffers from poor sample efficiency due to high-dimensional observations in complex tasks. While existing works have shown that vision-language models (VLMs) can assist RL, they often focus on knowledge distillation from the VLM to RL, overlooking the potential of
Conversational search aims to satisfy users’ complex information needs via multiple-turn interactions. The key challenge lies in revealing real users’ search intent from the context-dependent queries. Previous studies achieve conversational search by fine-tuning a conversational dense retriever with
Most reinforcement learning (RL) baselines maximize future cumulative rewards with a fixed single discount factor, which limits their performance in complex sequential decision-making tasks due to a failure to balance short-term objectives and long-term planning. To address this issue, this paper fo…
Recurrent large language models (Recurrent LLMs) offer linear computational complexity as efficient alternatives to quadratic self-attention-based LLMs (Self-Attention LLMs). However, Recurrent LLMs underperform on long-context tasks due to limited fixed-size memory. Previous research focused on arc…
Offline meta-reinforcement learning requires agents to generalize to unseen tasks from fixed datasets, yet existing sequence-based and MoE-based methods rely on implicit or token-level routing signals that fail to capture task-level structure. We propose the **Task-Guided Router (TGR)**, a structure…
The existing multi-atlas brain network analysis methods rely on some simple fusion methods (i.e., add and concatenation) and do not consider the information redundancy caused by increased brain regions. To improve upon these, we propose the Sparse and Uncertain Fusion Transformers (SUFT) for multi-a…
Conversational search facilitates complex information retrieval by enabling multi-turn interactions between users and the system. Supporting such interactions requires a comprehensive understanding of the conversational inputs to formulate a good search query based on historical information. In part…
Given the increasing number of parameter-efficient adapters of large language models (LLMs), how can we reuse them to improve LLM performance on new tasks? We study how to best build a *library* of adapters given multi-task data and devise techniques for both *zero-shot* and *supervised* task genera…
Parameter-efficient fine-tuning (PEFT) for cross-task generalization consists in pre-training adapters on a multi-task training set before few-shot adaptation to test tasks. Polytropon [Ponti et al., 2023] ($\texttt{Poly}$) jointly learns an inventory of adapters and a *routing* function that select…