The Inherent Limits of Pretrained LLMs: The Unexpected Convergence of Instruction Tuning and In-Context Learning Capabilities

Large Language Models (LLMs), trained on extensive web-scale corpora, have demonstrated remarkable abilities across diverse tasks, especially as they are scaled up. Nevertheless, even state-of-the-art models struggle in certain cases, sometimes failing at problems solvable by young children, indicating that traditional notions of task complexity are insufficient for explaining LLM capabilities. However, exploring LLM capabilities is complicated by the fact that most widely-used models are also "instruction-tuned" to respond appropriately to prompts. With the goal of disentangling the factors influencing LLM performance, we investigate whether instruction-tuned models possess fundamentally different capabilities from base models that are prompted using in-context examples. Through extensive experiments across various model families, scales and task types, which included instruction tuning 90 different LLMs, we demonstrate that the performance of instruction-tuned models is significantly correlated with the in-context performance of their base counterparts. By clarifying what instruction-tuning contributes, we extend prior research into in-context learning, which suggests that base models use priors from pretraining data to solve tasks. Specifically, we extend this understanding to instruction-tuned models, suggesting that their pretraining data similarly sets a limiting boundary on the tasks they can solve, with the added influence of the instruction-tuning dataset.

Adapting Whisper for Regional Dialects: Enhancing Public Services for Vulnerable Populations in the United Kingdom

We collect novel data in the public service domain to evaluate the capability of the state-of-the-art automatic speech recognition (ASR) models in capturing regional differences in accents in the United Kingdom (UK), specifically focusing on two accents from Scotland with distinct dialects. This study addresses real-world problems where biased ASR models can lead to miscommunication in public services, disadvantaging individuals with regional accents particularly those in vulnerable populations. We first examine the out-of-the-box performance of the Whisper large-v3 model on a baseline dataset and our data. We then explore the impact of fine-tuning Whisper on the performance in the two UK regions and investigate the effectiveness of existing model evaluation techniques for our real-world application through manual inspection of model errors. We observe that the Whisper model has a higher word error rate (WER) on our test datasets compared to the baseline data and fine-tuning on a given data improves performance on the test dataset with the same domain and accent. The fine-tuned models also appear to show improved performance when applied to the test data outside of the region it was trained on suggesting that fine-tuned models may be transferable within parts of the UK. Our manual analysis of model outputs reveals the benefits and drawbacks of using WER as an evaluation metric and fine-tuning to adapt to regional dialects.

Assessing Language Comprehension in Large Language Models Using Construction Grammar

Large Language Models, despite their significant capabilities, are known to fail in surprising and unpredictable ways. Evaluating their true `understanding' of language is particularly challenging due to the extensive web-scale data they are trained on. Therefore, we construct an evaluation to systematically assess natural language understanding (NLU) in LLMs by leveraging Construction Grammar (CxG), which provides insights into the meaning captured by linguistic elements known as constructions (Cxns). CxG is well-suited for this purpose because provides a theoretical basis to construct targeted evaluation sets. These datasets are carefully constructed to include examples which are unlikely to appear in pre-training data, yet intuitive and easy for humans to understand, enabling a more targeted and reliable assessment. Our experiments focus on downstream natural language inference and reasoning tasks by comparing LLMs' understanding of the underlying meanings communicated through 8 unique Cxns with that of humans. The results show that while LLMs demonstrate some knowledge of constructional information, even the latest models including GPT-o1 struggle with abstract meanings conveyed by these Cxns, as demonstrated in cases where test sentences are dissimilar to their pre-training data. We argue that such cases provide a more accurate test of true language understanding, highlighting key limitations in LLMs' semantic capabilities. We make our novel dataset and associated experimental data including prompts and model responses publicly available.

SpeciaLex: A Benchmark for In-Context Specialized Lexicon Learning

Specialized lexicons are collections of words with associated constraints such as special definitions, specific roles, and intended target audiences. These constraints are necessary for content generation and documentation tasks (e.g., writing technical manuals or children's books), where the goal is to reduce the ambiguity of text content and increase its overall readability for a specific group of audience. Understanding how large language models can capture these constraints can help researchers build better, more impactful tools for wider use beyond the NLP community. Towards this end, we introduce SpeciaLex, a benchmark for evaluating a language model's ability to follow specialized lexicon-based constraints across 18 diverse subtasks with 1,285 test instances covering core tasks of Checking, Identification, Rewriting, and Open Generation. We present an empirical evaluation of 15 open and closed-source LLMs and discuss insights on how factors such as model scale, openness, setup, and recency affect performance upon evaluating with the benchmark.

Fine-Tuning with Divergent Chains of Thought Boosts Reasoning Through Self-Correction in Language Models

Requiring a Large Language Model to generate intermediary reasoning steps has been shown to be an effective way of boosting performance. In fact, it has been found that instruction tuning on these intermediary reasoning steps improves model performance. In this work, we present a novel method of further improving performance by requiring models to compare multiple reasoning chains before generating a solution in a single inference step. We call this method Divergent CoT (DCoT). We find that instruction tuning on DCoT datasets boosts the performance of even smaller, and therefore more accessible, LLMs. Through a rigorous set of experiments spanning a wide range of tasks that require various reasoning types, we show that fine-tuning on DCoT consistently improves performance over the CoT baseline across model families and scales (1.3B to 70B). Through a combination of empirical and manual evaluation, we additionally show that these performance gains stem from models generating multiple divergent reasoning chains in a single inference step, indicative of the enabling of self-correction in language models. Our code and data are publicly available at https://github.com/UKPLab/arxiv2024-divergent-cot.

FS-RAG: A Frame Semantics Based Approach for Improved Factual Accuracy in Large Language Models

We present a novel extension to Retrieval Augmented Generation with the goal of mitigating factual inaccuracies in the output of large language models. Specifically, our method draws on the cognitive linguistic theory of frame semantics for the indexing and retrieval of factual information relevant to helping large language models answer queries. We conduct experiments to demonstrate the effectiveness of this method both in terms of retrieval effectiveness and in terms of the relevance of the frames and frame relations automatically generated. Our results show that this novel mechanism of Frame Semantic-based retrieval, designed to improve Retrieval Augmented Generation (FS-RAG), is effective and offers potential for providing data-driven insights into frame semantics theory. We provide open access to our program code and prompts.

Pre-Trained Language Models Represent Some Geographic Populations Better Than Others

This paper measures the skew in how well two families of LLMs represent diverse geographic populations. A spatial probing task is used with geo-referenced corpora to measure the degree to which pre-trained language models from the OPT and BLOOM series represent diverse populations around the world. Results show that these models perform much better for some populations than others. In particular, populations across the US and the UK are represented quite well while those in South and Southeast Asia are poorly represented. Analysis shows that both families of models largely share the same skew across populations. At the same time, this skew cannot be fully explained by sociolinguistic factors, economic factors, or geographic factors. The basic conclusion from this analysis is that pre-trained models do not equally represent the world's population: there is a strong skew towards specific geographic populations. This finding challenges the idea that a single model can be used for all populations.

Code-Mixed Probes Show How Pre-Trained Models Generalise On Code-Switched Text

Code-switching is a prevalent linguistic phenomenon in which multilingual individuals seamlessly alternate between languages. Despite its widespread use online and recent research trends in this area, research in code-switching presents unique challenges, primarily stemming from the scarcity of labelled data and available resources. In this study we investigate how pre-trained Language Models handle code-switched text in three dimensions: a) the ability of PLMs to detect code-switched text, b) variations in the structural information that PLMs utilise to capture code-switched text, and c) the consistency of semantic information representation in code-switched text. To conduct a systematic and controlled evaluation of the language models in question, we create a novel dataset of well-formed naturalistic code-switched text along with parallel translations into the source languages. Our findings reveal that pre-trained language models are effective in generalising to code-switched text, shedding light on the abilities of these models to generalise representations to CS corpora. We release all our code and data including the novel corpus at https://github.com/francesita/code-mixed-probes.

Standardize: Aligning Language Models with Expert-Defined Standards for Content Generation

Domain experts across engineering, healthcare, and education follow strict standards for producing quality content such as technical manuals, medication instructions, and children's reading materials. However, current works in controllable text generation have yet to explore using these standards as references for control. Towards this end, we introduce Standardize, a retrieval-style in-context learning-based framework to guide large language models to align with expert-defined standards. Focusing on English language standards in the education domain as a use case, we consider the Common European Framework of Reference for Languages (CEFR) and Common Core Standards (CCS) for the task of open-ended content generation. Our findings show that models can gain 40% to 100% increase in precise accuracy for Llama2 and GPT-4, respectively, demonstrating that the use of knowledge artifacts extracted from standards and integrating them in the generation process can effectively guide models to produce better standard-aligned content.

Word Boundary Information Isn't Useful for Encoder Language Models

All existing transformer-based approaches to NLP using subword tokenisation algorithms encode whitespace (word boundary information) through the use of special space symbols (such as \#\# or \_) forming part of tokens. These symbols have been shown to a) lead to reduced morphological validity of tokenisations, and b) give substantial vocabulary redundancy. As such, removing these symbols has been shown to have a beneficial effect on the processing of morphologically complex words for transformer encoders in the pretrain-finetune paradigm. In this work, we explore whether word boundary information is at all useful to such models. In particular, we train transformer encoders across four different training scales, and investigate several alternative approaches to including word boundary information, evaluating on a range of tasks across different domains and problem set-ups: GLUE (for sentence-level classification), NER (for token-level classification), and two classification datasets involving complex words (Superbizarre and FLOTA). Overall, through an extensive experimental setup that includes the pre-training of 29 models, we find no substantial improvements from our alternative approaches, suggesting that modifying tokenisers to remove word boundary information isn't leading to a loss of useful information.