The advent of large language models (LLMs) has enabled significant performance gains in the field of natural language processing. However, recent studies have found that LLMs often resort to shortcuts when performing tasks, creating an illusion of enhanced performance while lacking generalizability in their decision rules. This phenomenon introduces challenges in accurately assessing natural language understanding in LLMs. Our paper provides a concise survey of relevant research in this area and puts forth a perspective on the implications of shortcut learning in the evaluation of language models, specifically for NLU tasks. This paper urges more research efforts to be put towards deepening our comprehension of shortcut learning, contributing to the development of more robust language models, and raising the standards of NLU evaluation in real-world scenarios.
相關內容
The growing integration of large language models (LLMs) into social operations amplifies their impact on decisions in crucial areas such as economics, law, education, and healthcare, raising public concerns about these models' discrimination-related safety and reliability. However, prior discrimination measuring frameworks solely assess the average discriminatory behavior of LLMs, often proving inadequate due to the overlook of an additional discrimination-leading factor, i.e., the LLMs' prediction variation across diverse contexts. In this work, we present the Prejudice-Caprice Framework (PCF) that comprehensively measures discrimination in LLMs by considering both their consistently biased preference and preference variation across diverse contexts. Specifically, we mathematically dissect the aggregated contextualized discrimination risk of LLMs into prejudice risk, originating from LLMs' persistent prejudice, and caprice risk, stemming from their generation inconsistency. In addition, we utilize a data-mining approach to gather preference-detecting probes from sentence skeletons, devoid of attribute indications, to approximate LLMs' applied contexts. While initially intended for assessing discrimination in LLMs, our proposed PCF facilitates the comprehensive and flexible measurement of any inductive biases, including knowledge alongside prejudice, across various modality models. We apply our discrimination-measuring framework to 12 common LLMs, yielding intriguing findings: i) modern LLMs demonstrate significant pro-male stereotypes, ii) LLMs' exhibited discrimination correlates with several social and economic factors, iii) prejudice risk dominates the overall discrimination risk and follows a normal distribution, and iv) caprice risk contributes minimally to the overall risk but follows a fat-tailed distribution, suggesting that it is wild risk requiring enhanced surveillance.
With the advent of large language models (LLMs) like GPT-3, a natural question is the extent to which these models can be utilized for source code optimization. This paper presents methodologically stringent case studies applied to well-known open source python libraries pillow and numpy. We find that contemporary LLM ChatGPT-4 (state September and October 2023) is surprisingly adept at optimizing energy and compute efficiency. However, this is only the case in interactive use, with a human expert in the loop. Aware of experimenter bias, we document our qualitative approach in detail, and provide transcript and source code. We start by providing a detailed description of our approach in conversing with the LLM to optimize the _getextrema function in the pillow library, and a quantitative evaluation of the performance improvement. To demonstrate qualitative replicability, we report further attempts on another locus in the pillow library, and one code locus in the numpy library, to demonstrate generalization within and beyond a library. In all attempts, the performance improvement is significant (factor up to 38). We have also not omitted reporting of failed attempts (there were none). We conclude that LLMs are a promising tool for code optimization in open source libraries, but that the human expert in the loop is essential for success. Nonetheless, we were surprised by how few iterations were required to achieve substantial performance improvements that were not obvious to the expert in the loop. We would like bring attention to the qualitative nature of this study, more robust quantitative studies would need to introduce a layer of selecting experts in a representative sample -- we invite the community to collaborate.
Existing research has shown that large language models (LLMs) exhibit remarkable performance in language understanding and generation. However, when LLMs are continuously fine-tuned on complex and diverse domain-specific downstream tasks, the inference performance on historical tasks decreases dramatically, which is known as a catastrophic forgetting problem. A trade-off needs to be kept between learning plasticity and memory stability. Plenty of existing works have explored strategies like memory replay, regularization and parameter isolation, but little is known about the geometric connection of various adjacent minima in the continual LLMs fine-tuning scenarios. In this work, we investigate the geometric connections of different minima through the lens of mode connectivity, which means different minima can be connected by a low-loss valley. Through extensive experiments, we uncover the mode connectivity phenomenon in the LLMs continual learning scenario and find that it can strike a balance between plasticity and stability. Building upon these findings, we propose a simple yet effective method called Interpolation-based LoRA (I-LoRA), which constructs a dual-memory experience replay framework based on LoRA parameter interpolations. Extensive experiments and analysis on eight domain-specific CL benchmarks demonstrate that I-LoRA consistently show significant improvement over the previous state-of-the-art approaches with up to $11\%$ performance gains, providing a strong baseline and insights for future research on the large language model continual learning problem. Our code is available at \url{//github.com/which47/LLMCL}.
Fine-grained control over large language models (LLMs) remains a significant challenge, hindering their adaptability to diverse user needs. While Reinforcement Learning from Human Feedback (RLHF) shows promise in aligning LLMs, its reliance on scalar rewards often limits its ability to capture diverse user preferences in real-world applications. To address this limitation, we introduce the Directional Preference Alignment (DPA) framework. Unlike the scalar-reward RLHF, DPA incorporates multi-objective reward modeling to represent diverse preference profiles. Additionally, DPA models user preferences as directions (i.e., unit vectors) in the reward space to achieve user-dependent preference control. Our method involves training a multi-objective reward model and then fine-tuning the LLM with a preference-conditioned variant of Rejection Sampling Finetuning (RSF), an RLHF method adopted by Llama 2. This method enjoys a better performance trade-off across various reward objectives. In comparison with the scalar-reward RLHF, DPA offers users intuitive control over LLM generation: they can arithmetically specify their desired trade-offs (e.g., more helpfulness with less verbosity). We also validate the effectiveness of DPA with real-world alignment experiments on Mistral-7B. Our method provides straightforward arithmetic control over the trade-off between helpfulness and verbosity while maintaining competitive performance with strong baselines such as Direct Preference Optimization (DPO).
As large language models increase in capability, researchers have started to conduct surveys of all kinds on these models in order to investigate the population represented by their responses. In this work, we critically examine language models' survey responses on the basis of the well-established American Community Survey by the U.S. Census Bureau and investigate whether they elicit a faithful representations of any human population. Using a de-facto standard multiple-choice prompting technique and evaluating 39 different language models using systematic experiments, we establish two dominant patterns: First, models' responses are governed by ordering and labeling biases, leading to variations across models that do not persist after adjusting for systematic biases. Second, models' responses do not contain the entropy variations and statistical signals typically found in human populations. As a result, a binary classifier can almost perfectly differentiate model-generated data from the responses of the U.S. census. At the same time, models' relative alignment with different demographic subgroups can be predicted from the subgroups' entropy, irrespective of the model's training data or training strategy. Taken together, our findings suggest caution in treating models' survey responses as equivalent to those of human populations.
Large language models (LLMs) are recognized as systems that closely mimic aspects of human intelligence. This capability has attracted attention from the social science community, who see the potential in leveraging LLMs to replace human participants in experiments, thereby reducing research costs and complexity. In this paper, we introduce a framework for large language models personification, including a strategy for constructing virtual characters' life stories from the ground up, a Multi-Agent Cognitive Mechanism capable of simulating human cognitive processes, and a psychology-guided evaluation method to assess human simulations from both self and observational perspectives. Experimental results demonstrate that our constructed simulacra can produce personified responses that align with their target characters. Our work is a preliminary exploration which offers great potential in practical applications. All the code and datasets will be released, with the hope of inspiring further investigations.
Aligning language models (LMs) with human opinion is challenging yet vital to enhance their grasp of human values, preferences, and beliefs. We present ChOiRe, a four-step framework to predict human opinion which differentially models the user explicit personae (i.e. demographic or ideological attributes) that are manually declared, and implicit personae inferred from user historical opinions. ChOiRe consists of (i) an LM analyzing the user explicit personae to filter out irrelevant attributes; (ii) the LM ranking the implicit persona opinions into a preferential list; (iii) Chain-of-Opinion (CoO) reasoning, where the LM sequentially analyzes the explicit personae and the most relevant implicit personae to perform opinion prediction; (iv) and where ChOiRe executes Step (iii) CoO multiple times with increasingly larger lists of implicit personae to overcome insufficient personae information to infer a final result. ChOiRe achieves new state-of-the-art effectiveness with limited inference calls, improving previous techniques significantly by 3.22%. We also show that ChOiRe Steps (i) and (ii) can significantly better fine-tune opinion-aligned models, by up to 18.44%.
Large language models (LLMs) have demonstrated impressive capabilities in natural language processing. However, their internal mechanisms are still unclear and this lack of transparency poses unwanted risks for downstream applications. Therefore, understanding and explaining these models is crucial for elucidating their behaviors, limitations, and social impacts. In this paper, we introduce a taxonomy of explainability techniques and provide a structured overview of methods for explaining Transformer-based language models. We categorize techniques based on the training paradigms of LLMs: traditional fine-tuning-based paradigm and prompting-based paradigm. For each paradigm, we summarize the goals and dominant approaches for generating local explanations of individual predictions and global explanations of overall model knowledge. We also discuss metrics for evaluating generated explanations, and discuss how explanations can be leveraged to debug models and improve performance. Lastly, we examine key challenges and emerging opportunities for explanation techniques in the era of LLMs in comparison to conventional machine learning models.
Large language models (LLMs) have significantly advanced the field of natural language processing (NLP), providing a highly useful, task-agnostic foundation for a wide range of applications. The great promise of LLMs as general task solvers motivated people to extend their functionality largely beyond just a ``chatbot'', and use it as an assistant or even replacement for domain experts and tools in specific domains such as healthcare, finance, and education. However, directly applying LLMs to solve sophisticated problems in specific domains meets many hurdles, caused by the heterogeneity of domain data, the sophistication of domain knowledge, the uniqueness of domain objectives, and the diversity of the constraints (e.g., various social norms, cultural conformity, religious beliefs, and ethical standards in the domain applications). To fill such a gap, explosively-increase research, and practices have been conducted in very recent years on the domain specialization of LLMs, which, however, calls for a comprehensive and systematic review to better summarizes and guide this promising domain. In this survey paper, first, we propose a systematic taxonomy that categorizes the LLM domain-specialization techniques based on the accessibility to LLMs and summarizes the framework for all the subcategories as well as their relations and differences to each other. We also present a comprehensive taxonomy of critical application domains that can benefit from specialized LLMs, discussing their practical significance and open challenges. Furthermore, we offer insights into the current research status and future trends in this area.
The emergence of large language models (LLMs) has substantially influenced natural language processing, demonstrating exceptional results across various tasks. In this study, we employ ``Introspective Tips" to facilitate LLMs in self-optimizing their decision-making. By introspectively examining trajectories, LLM refines its policy by generating succinct and valuable tips. Our method enhances the agent's performance in both few-shot and zero-shot learning situations by considering three essential scenarios: learning from the agent's past experiences, integrating expert demonstrations, and generalizing across diverse games. Importantly, we accomplish these improvements without fine-tuning the LLM parameters; rather, we adjust the prompt to generalize insights from the three aforementioned situations. Our framework not only supports but also emphasizes the advantage of employing LLM in in-contxt decision-making. Experiments involving over 100 games in TextWorld illustrate the superior performance of our approach.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191