Related papers: Linguistic communication as (inverse) reward desig…

Natural and formal languages provide an effective mechanism for humans to specify instructions and reward functions. We investigate how to generate policies via RL when reward functions are specified in a symbolic language captured by…

We show that explicit pragmatic inference aids in correctly generating and following natural language instructions for complex, sequential tasks. Our pragmatics-enabled models reason about why speakers produce certain instructions, and…

A distinguishing property of human intelligence is the ability to flexibly use language in order to communicate complex ideas with other humans in a variety of contexts. Research in natural language dialogue should focus on designing…

Speakers communicate to influence their partner's beliefs and shape their actions. Belief- and action-based objectives have been explored independently in recent computational models, but it has been challenging to explicitly compare or…

Learning from human feedback has gained traction in fields like robotics and natural language processing in recent years. While prior works mostly rely on human feedback in the form of comparisons, language is a preferable modality that…

End-to-end models for goal-orientated dialogue are challenging to train, because linguistic and strategic aspects are entangled in latent state vectors. We introduce an approach to learning representations of messages in dialogues by…

Language models are capable of iteratively improving their outputs based on natural language feedback, thus enabling in-context optimization of user preference. In place of human users, a second language model can be used as an evaluator,…

A number of recent works have proposed techniques for end-to-end learning of communication protocols among cooperative multi-agent populations, and have simultaneously found the emergence of grounded human-interpretable language in the…

Designing effective reward functions is crucial to training reinforcement learning (RL) algorithms. However, this design is non-trivial, even for domain experts, due to the subjective nature of certain tasks that are hard to quantify…

Prediction in language has traditionally been studied using simple designs in which neural responses to expected and unexpected words are compared in a categorical fashion. However, these designs have been contested as being `prediction…

We consider a finite-horizon discrete-time dynamic system jointly controlled by a designer and one or more agents, where the designer can influence the agents' actions through selective information disclosure. At each time step, the…

Reward design is a fundamental problem in reinforcement learning (RL). A misspecified or poorly designed reward can result in low sample efficiency and undesired behaviors. In this paper, we propose the idea of programmatic reward design,…

Natural language provides a widely accessible and expressive interface for robotic agents. To understand language in complex environments, agents must reason about the full range of language inputs and their correspondence to the world.…

Navigation guided by natural language instructions presents a challenging reasoning problem for instruction followers. Natural language instructions typically identify only a few high-level decisions and landmarks rather than complete…

A centerpiece of the ever-popular reinforcement learning from human feedback (RLHF) approach to fine-tuning autoregressive language models is the explicit training of a reward model to emulate human feedback, distinct from the language…

We are increasingly surrounded by artificially intelligent technology that takes decisions and executes actions on our behalf. This creates a pressing need for general means to communicate with, instruct and guide artificial agents, with…

Linguistic relations in oral conversations present how opinions are constructed and developed in a restricted time. The relations bond ideas, arguments, thoughts, and feelings, re-shape them during a speech, and finally build knowledge out…

Reinforcement learning (RL), particularly in sparse reward settings, often requires prohibitively large numbers of interactions with the environment, thereby limiting its applicability to complex problems. To address this, several prior…

To align conditional text generation model outputs with desired behaviors, there has been an increasing focus on training the model using reinforcement learning (RL) with reward functions learned from human annotations. Under this…

Reward Models (RMs) are crucial for online alignment of language models (LMs) with human preferences. However, RM-based preference-tuning is vulnerable to reward hacking, whereby LM policies learn undesirable behaviors from flawed RMs. By…