Improving Context-Aware Preference Modeling for Language Models

While finetuning language models from pairwise preferences has proven remarkably effective, the underspecified nature of natural language presents critical challenges. Direct preference feedback is uninterpretable, difficult to provide where multidimensional criteria may apply, and often inconsistent, either because it is based on incomplete instructions or provided by diverse principals. To address these challenges, we consider the two-step preference modeling procedure that first resolves the under-specification by selecting a context, and then evaluates preference with respect to the chosen context. We decompose reward modeling error according to these two steps, which suggests that supervising context in addition to context-specific preference may be a viable approach to aligning models with diverse human preferences. For this to work, the ability of models to evaluate context-specific preference is critical. To this end, we contribute context-conditioned preference datasets and accompanying experiments that investigate the ability of language models to evaluate context-specific preference. We use our datasets to (1) show that existing preference models benefit from, but fail to fully consider, added context, (2) finetune a context-aware reward model with context-specific performance exceeding that of GPT-4 and Llama 3 70B on tested datasets, and (3) investigate the value of context-aware preference modeling.

Faux Polyglot: A Study on Information Disparity in Multilingual Large Language Models

With Retrieval Augmented Generation (RAG), Large Language Models (LLMs) are playing a pivotal role in information search and are being adopted globally. Although the multilingual capability of LLMs offers new opportunities to bridge the language barrier, do these capabilities translate into real-life scenarios where linguistic divide and knowledge conflicts between multilingual sources are known occurrences? In this paper, we studied LLM's linguistic preference in a RAG-based information search setting. We found that LLMs displayed systemic bias towards information in the same language as the query language in both information retrieval and answer generation. Furthermore, in scenarios where there is little information in the language of the query, LLMs prefer documents in high-resource languages, reinforcing the dominant views. Such bias exists for both factual and opinion-based queries. Our results highlight the linguistic divide within multilingual LLMs in information search systems. The seemingly beneficial multilingual capability of LLMs may backfire on information parity by reinforcing language-specific information cocoons or filter bubbles further marginalizing low-resource views.

ECBD: Evidence-Centered Benchmark Design for NLP

Benchmarking is seen as critical to assessing progress in NLP. However, creating a benchmark involves many design decisions (e.g., which datasets to include, which metrics to use) that often rely on tacit, untested assumptions about what the benchmark is intended to measure or is actually measuring. There is currently no principled way of analyzing these decisions and how they impact the validity of the benchmark's measurements. To address this gap, we draw on evidence-centered design in educational assessments and propose Evidence-Centered Benchmark Design (ECBD), a framework which formalizes the benchmark design process into five modules. ECBD specifies the role each module plays in helping practitioners collect evidence about capabilities of interest. Specifically, each module requires benchmark designers to describe, justify, and support benchmark design choices -- e.g., clearly specifying the capabilities the benchmark aims to measure or how evidence about those capabilities is collected from model responses. To demonstrate the use of ECBD, we conduct case studies with three benchmarks: BoolQ, SuperGLUE, and HELM. Our analysis reveals common trends in benchmark design and documentation that could threaten the validity of benchmarks' measurements.

Can Language Models Serve as Text-Based World Simulators?

Virtual environments play a key role in benchmarking advances in complex planning and decision-making tasks but are expensive and complicated to build by hand. Can current language models themselves serve as world simulators, correctly predicting how actions change different world states, thus bypassing the need for extensive manual coding? Our goal is to answer this question in the context of text-based simulators. Our approach is to build and use a new benchmark, called ByteSized32-State-Prediction, containing a dataset of text game state transitions and accompanying game tasks. We use this to directly quantify, for the first time, how well LLMs can serve as text-based world simulators. We test GPT-4 on this dataset and find that, despite its impressive performance, it is still an unreliable world simulator without further innovations. This work thus contributes both new insights into current LLM's capabilities and weaknesses, as well as a novel benchmark to track future progress as new models appear.

Generative Echo Chamber? Effects of LLM-Powered Search Systems on Diverse Information Seeking

Large language models (LLMs) powered conversational search systems have already been used by hundreds of millions of people, and are believed to bring many benefits over conventional search. However, while decades of research and public discourse interrogated the risk of search systems in increasing selective exposure and creating echo chambers -- limiting exposure to diverse opinions and leading to opinion polarization, little is known about such a risk of LLM-powered conversational search. We conduct two experiments to investigate: 1) whether and how LLM-powered conversational search increases selective exposure compared to conventional search; 2) whether and how LLMs with opinion biases that either reinforce or challenge the user's view change the effect. Overall, we found that participants engaged in more biased information querying with LLM-powered conversational search, and an opinionated LLM reinforcing their views exacerbated this bias. These results present critical implications for the development of LLMs and conversational search systems, and the policy governing these technologies.

Deep Language Networks: Joint Prompt Training of Stacked LLMs using Variational Inference

We view large language models (LLMs) as stochastic \emph{language layers} in a network, where the learnable parameters are the natural language \emph{prompts} at each layer. We stack two such layers, feeding the output of one layer to the next. We call the stacked architecture a \emph{Deep Language Network} (DLN). We first show how to effectively perform prompt optimization for a 1-Layer language network (DLN-1). We then show how to train 2-layer DLNs (DLN-2), where two prompts must be learnt. We consider the output of the first layer as a latent variable to marginalize, and devise a variational inference algorithm for joint prompt training. A DLN-2 reaches higher performance than a single layer, sometimes comparable to few-shot GPT-4 even when each LLM in the network is smaller and less powerful. The DLN code is open source: https://github.com/microsoft/deep-language-networks .

Rethinking Model Evaluation as Narrowing the Socio-Technical Gap

The recent development of generative and large language models (LLMs) poses new challenges for model evaluation that the research community and industry are grappling with. While the versatile capabilities of these models ignite excitement, they also inevitably make a leap toward homogenization: powering a wide range of applications with a single, often referred to as ``general-purpose'', model. In this position paper, we argue that model evaluation practices must take on a critical task to cope with the challenges and responsibilities brought by this homogenization: providing valid assessments for whether and how much human needs in downstream use cases can be satisfied by the given model (\textit{socio-technical gap}). By drawing on lessons from the social sciences, human-computer interaction (HCI), and the interdisciplinary field of explainable AI (XAI), we urge the community to develop evaluation methods based on real-world socio-requirements and embrace diverse evaluation methods with an acknowledgment of trade-offs between realism to socio-requirements and pragmatic costs. By mapping HCI and current NLG evaluation methods, we identify opportunities for new evaluation methods for LLMs to narrow the socio-technical gap and pose open questions.

ByteSized32: A Corpus and Challenge Task for Generating Task-Specific World Models Expressed as Text Games

In this work we examine the ability of language models to generate explicit world models of scientific and common-sense reasoning tasks by framing this as a problem of generating text-based games. To support this, we introduce ByteSized32, a corpus of 32 highly-templated text games written in Python totaling 24k lines of code, each centered around a particular task, and paired with a set of 16 unseen text game specifications for evaluation. We propose a suite of automatic and manual metrics for assessing simulation validity, compliance with task specifications, playability, winnability, and alignment with the physical world. In a single-shot evaluation of GPT-4 on this simulation-as-code-generation task, we find it capable of producing runnable games in 27% of cases, highlighting the difficulty of this challenge task. We discuss areas of future improvement, including GPT-4's apparent capacity to perform well at simulating near canonical task solutions, with performance dropping off as simulations include distractors or deviate from canonical solutions in the action space.

Evaluating NLG Evaluation Metrics: A Measurement Theory Perspective

We address the fundamental challenge in Natural Language Generation (NLG) model evaluation, the design and validation of evaluation metrics. Recognizing the limitations of existing metrics and issues with human judgment, we propose using measurement theory, the foundation of test design, as a framework for conceptualizing and evaluating the validity and reliability of NLG evaluation metrics. This approach offers a systematic method for defining "good" metrics, developing robust metrics, and assessing metric performance. In this paper, we introduce core concepts in measurement theory in the context of NLG evaluation and key methods to evaluate the performance of NLG metrics. Through this framework, we aim to promote the design, evaluation, and interpretation of valid and reliable metrics, ultimately contributing to the advancement of robust and effective NLG models in real-world settings.

Supporting Qualitative Analysis with Large Language Models: Combining Codebook with GPT-3 for Deductive Coding

Qualitative analysis of textual contents unpacks rich and valuable information by assigning labels to the data. However, this process is often labor-intensive, particularly when working with large datasets. While recent AI-based tools demonstrate utility, researchers may not have readily available AI resources and expertise, let alone be challenged by the limited generalizability of those task-specific models. In this study, we explored the use of large language models (LLMs) in supporting deductive coding, a major category of qualitative analysis where researchers use pre-determined codebooks to label the data into a fixed set of codes. Instead of training task-specific models, a pre-trained LLM could be used directly for various tasks without fine-tuning through prompt learning. Using a curiosity-driven questions coding task as a case study, we found, by combining GPT-3 with expert-drafted codebooks, our proposed approach achieved fair to substantial agreements with expert-coded results. We lay out challenges and opportunities in using LLMs to support qualitative coding and beyond.