Skip to main content ->
Ai2

Research - Papers

Explore a selection of our published work on a variety of key research challenges in AI.

Filter papers

Are Machine Rationales (Not) Useful to Humans? Measuring and Improving Human Utility of Free-Text Rationales

Brihi JoshiZiyi LiuSahana RamnathXiang Ren
2023
arXiv.org

Among the remarkable emergent capabilities of large language models (LMs) is free-text rationalization; beyond a certain scale, large LMs are capable of generating seemingly useful rationalizations,… 

Embedding Recycling for Language Models

Jon Saad-FalconAmanpreet SinghLuca SoldainiDoug Downey
2023
Findings of EACL

Training and inference with large neural models is expensive. However, for many application domains, while new tasks and models arise frequently, the underlying doc-uments being modeled remain… 

ArK: Augmented Reality with Knowledge Interactive Emergent Ability

Qiuyuan HuangJ. ParkAbhinav GuptaJianfeng Gao
2023
arXiv.org

Despite the growing adoption of mixed reality and interactive AI agents, it remains challenging for these systems to generate high quality 2D/3D scenes in unseen environments. The common practice… 

Binding Language Models in Symbolic Languages

Zhoujun ChengTianbao XiePeng ShiTao Yu
2023
ICLR • Proceedings

Though end-to-end neural approaches have recently been dominating NLP tasks in both performance and ease-of-use, they lack interpretability and robustness. We propose Binder, a training-free… 

Can AI language models replace human participants?

Danica DillionNiket TandonYuling GuKurt Gray
2023
Trends in Cognitive Sciences

Recent work suggests that language models such as GPT can make human-like judgments across a number of domains. We explore whether and when language models might replace human participants in… 

Complexity-Based Prompting for Multi-Step Reasoning

Yao FuHao PengAshish SabharwalTushar Khot
2023
ICLR

We study the task of prompting large-scale language models to perform multi-step reasoning. Existing work shows that when prompted with a chain of thoughts (CoT), sequences of short sentences… 

Decomposed Prompting: A Modular Approach for Solving Complex Tasks

Tushar KhotHarsh TrivediMatthew FinlaysonAshish Sabharwal
2023
ICLR

Few-shot prompting is a surprisingly powerful way to use Large Language Models (LLMs) to solve various tasks. However, this approach struggles as the task complexity increases or when the individual… 

Editing Models with Task Arithmetic

Gabriel IlharcoMarco Tulio RibeiroMitchell WortsmanAli Farhadi
2023
ICLR

Changing how pre-trained models behave -- e.g., improving their performance on a downstream task or mitigating biases learned during pre-training -- is a common practice when developing machine… 

InSCIt: Information-Seeking Conversations with Mixed-Initiative Interactions

Zeqiu WuRyu ParishHao ChengHannaneh Hajishirzi
2023
TACL

In an information-seeking conversation, a user may ask questions that are under-specified or unanswerable. An ideal agent would interact by initiating different response types according to the… 

Is Reinforcement Learning (Not) for Natural Language Processing?: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization

Rajkumar RamamurthyPrithviraj AmmanabroluKianté BrantleyYejin Choi
2023
ICLR

We tackle the problem of aligning pre-trained large language models (LMs) with human preferences. If we view text generation as a sequential decision-making problem, reinforcement learning (RL)…