Li Zhang

Sebastian Gehrmann, Abhik Bhattacharjee, Abinaya Mahendiran et al. · amazon-science, cmu

Evaluation in machine learning is usually informed by past choices, for example which datasets or metrics to use. This standardization enables the comparison on equal footing using leaderboards, but the evaluation choices become sub-optimal as better alternatives arise. This problem is especially pertinent in natural language generation which requires ever-improving suites of datasets, metrics, and human evaluation to make definitive claims. To make following best model evaluation practices easier, we introduce GEMv2. The new version of the Generation, Evaluation, and Metrics Benchmark introduces a modular infrastructure for dataset, model, and metric developers to benefit from each others work. GEMv2 supports 40 documented datasets in 51 languages. Models for all datasets can be evaluated online and our interactive data card creation and rendering tools make it easier to add new datasets to the living benchmark.

Li Zhang, Hainiu Xu, Yue Yang et al. · allen-ai, cmu

Entities and events are crucial to natural language reasoning and common in procedural texts. Existing work has focused either exclusively on entity state tracking (e.g., whether a pan is hot) or on event reasoning (e.g., whether one would burn themselves by touching the pan), while these two tasks are often causally related. We propose CREPE, the first benchmark on causal reasoning of event plausibility and entity states. We show that most language models, including GPT-3, perform close to chance at .35 F1, lagging far behind human at .87 F1. We boost model performance to .59 F1 by creatively representing events as programming languages while prompting language models pretrained on code. By injecting the causal relations between entities and events as intermediate reasoning steps in our representation, we further boost the performance to .67 F1. Our findings indicate not only the challenge that CREPE brings for language models, but also the efficacy of code-like prompting combined with chain-of-thought prompting for multihop event reasoning.

Qing Lyu, Shreya Havaldar, Adam Stein et al.

While Chain-of-Thought (CoT) prompting boosts Language Models' (LM) performance on a gamut of complex reasoning tasks, the generated reasoning chain does not necessarily reflect how the model arrives at the answer (aka. faithfulness). We propose Faithful CoT, a reasoning framework involving two stages: Translation (Natural Language query $\rightarrow$ symbolic reasoning chain) and Problem Solving (reasoning chain $\rightarrow$ answer), using an LM and a deterministic solver respectively. This guarantees that the reasoning chain provides a faithful explanation of the final answer. Aside from interpretability, Faithful CoT also improves empirical performance: it outperforms standard CoT on 9 of 10 benchmarks from 4 diverse domains, with a relative accuracy gain of 6.3% on Math Word Problems (MWP), 3.4% on Planning, 5.5% on Multi-hop Question Answering (QA), and 21.4% on Relational Inference. Furthermore, with GPT-4 and Codex, it sets the new state-of-the-art few-shot performance on 7 datasets (with 95.0+ accuracy on 6 of them), showing a strong synergy between faithfulness and accuracy.

Bodhisattwa Prasad Majumder, Bhavana Dalvi Mishra, Peter Jansen et al.

Language agents have shown some ability to interact with an external environment, e.g., a virtual world such as ScienceWorld, to perform complex tasks, e.g., growing a plant, without the startup costs of reinforcement learning. However, despite their zero-shot capabilities, these agents to date do not continually improve over time beyond performance refinement on a specific task. Here we present CLIN, the first language-based agent to achieve this, so that it continually improves over multiple trials, including when both the environment and task are varied, and without requiring parameter updates. Our approach is to use a persistent, dynamic, textual memory centered on causal abstractions (rather than general "helpful hints") that is regularly updated after each trial so that the agent gradually learns useful knowledge for new trials. In the ScienceWorld benchmark, CLIN is able to continually improve on repeated trials on the same task and environment, outperforming state-of-the-art reflective language agents like Reflexion by 23 absolute points. CLIN can also transfer its learning to new environments (or new tasks), improving its zero-shot performance by 4 points (13 for new tasks) and can further improve performance there through continual memory updates, enhancing performance by an additional 17 points (7 for new tasks). This suggests a new architecture for agents built on frozen models that can still continually and rapidly improve over time.

Li Zhang, Liam Dugan, Hainiu Xu et al.

Recent work has shown that prompting language models with code-like representations of natural language leads to performance improvements on structured reasoning tasks. However, such tasks comprise only a small subset of all natural language tasks. In our work, we seek to answer whether or not code-prompting is the preferred way of interacting with language models in general. We compare code and text prompts across three popular GPT models (davinci, code-davinci-002, and text-davinci-002) on a broader selection of tasks (e.g., QA, sentiment, summarization) and find that with few exceptions, code prompts do not consistently outperform text prompts. Furthermore, we show that the style of code prompt has a large effect on performance for some but not all tasks and that fine-tuning on text instructions leads to better relative performance of code prompts.

Zhaoyi Joey Hou, Li Zhang, Chris Callison-Burch

Script learning studies how stereotypical events unfold, enabling machines to reason about narratives with implicit information. Previous works mostly consider a script as a linear sequence of events while ignoring the potential branches that arise due to people's circumstantial choices. We hence propose Choice-75, the first benchmark that challenges intelligent systems to make decisions given descriptive scenarios, containing 75 scripts and more than 600 scenarios. We also present preliminary results with current large language models (LLM). Although they demonstrate overall decent performance, there is still notable headroom in hard scenarios.

Fang Liu, Jia Li, Li Zhang

There is growing interest in software migration as the development of software and society. Manually migrating projects between languages is error-prone and expensive. In recent years, researchers have begun to explore automatic program translation using supervised deep learning techniques by learning from large-scale parallel code corpus. However, parallel resources are scarce in the programming language domain, and it is costly to collect bilingual data manually. To address this issue, several unsupervised programming translation systems are proposed. However, these systems still rely on huge monolingual source code to train, which is very expensive. Besides, these models cannot perform well for translating the languages that are not seen during the pre-training procedure. In this paper, we propose SDA-Trans, a syntax and domain-aware model for program translation, which leverages the syntax structure and domain knowledge to enhance the cross-lingual transfer ability. SDA-Trans adopts unsupervised training on a smaller-scale corpus, including Python and Java monolingual programs. The experimental results on function translation tasks between Python, Java, and C++ show that SDA-Trans outperforms many large-scale pre-trained models, especially for unseen language translation.

Donghao Yang, Aolang Wu, Tianyi Zhang et al.

Among the programming languages for Programmable Logic Controllers (PLCs), Structured Text (ST) is widely adopted for industrial automation due to its expressiveness and flexibility. However, major vendors implement ST with proprietary extensions and hardware-specific libraries - Siemens' SCL and CODESYS' ST each differ in syntax and functionality. This fragmentation forces engineers to relearn implementation details across platforms, creating substantial productivity barriers. To address this challenge, we developed AutoPLC, a framework capable of automatically generating vendor-aware ST code directly from natural language requirements. Our solution begins by building two essential knowledge sources tailored to each vendor's specifications: a structured API library containing platform-exclusive functions, and an annotated case database that captures real-world implementation experience. Building on these foundations, we created a four-stage generation process that combines step-wise planning (enhanced with a lightweight natural language state machine support for control logic), contextual case retrieval using LLM-based reranking, API recommendation guided by industrial data, and dynamic validation through direct interaction with vendor IDEs. Implemented for Siemens TIA Portal and the CODESYS platform, AutoPLC achieves 90%+ compilation success on our 914-task benchmark (covering general-purpose and process control functions), outperforming all selected baselines, at an average cost of only $0.13 per task. Experienced PLC engineers positively assessed the practical utility of the generated code, including cases that failed compilation. We open-source our framework at https://github.com/cangkui/AutoPLC.

Xuecheng Zou, Ke Liu, Bingbing Wang et al.

Building upon the standard graph-based Retrieval-Augmented Generation (RAG), the introduction of heterogeneous graphs and hypergraphs aims to enrich retrieval and generation by leveraging the relationships between multiple entities through the concept of semantic units (SUs). But this also raises a key issue: The extraction of high-level SUs limited to local text chunks is prone to ambiguity, complex coupling, and increased retrieval overhead due to the lack of global knowledge or the neglect of fine-grained relationships. To address these issues, we propose GOSU, a semantic unit-centric RAG framework that efficiently performs global disambiguation and utilizes SUs to capture interconnections between different nodes across the global context. In the graph construction phase, GOSU performs global merging on the pre-extracted SUs from local text chunks and guides entity and relationship extraction, reducing the difficulty of coreference resolution while uncovering global semantic objects across text chunks. In the retrieval and generation phase, we introduce hierarchical keyword extraction and semantic unit completion. The former uncovers the fine-grained binary relationships overlooked by the latter, while the latter compensates for the coarse-grained n-ary relationships missing from the former. Evaluation across multiple tasks demonstrates that GOSU outperforms the baseline RAG methods in terms of generation quality.

Mengwei Xu, Wangsong Yin, Dongqi Cai et al.

Large foundation models, including large language models (LLMs), vision transformers (ViTs), diffusion, and LLM-based multimodal models, are revolutionizing the entire machine learning lifecycle, from training to deployment. However, the substantial advancements in versatility and performance these models offer come at a significant cost in terms of hardware resources. To support the growth of these large models in a scalable and environmentally sustainable way, there has been a considerable focus on developing resource-efficient strategies. This survey delves into the critical importance of such research, examining both algorithmic and systemic aspects. It offers a comprehensive analysis and valuable insights gleaned from existing literature, encompassing a broad array of topics from cutting-edge model architectures and training/serving algorithms to practical system designs and implementations. The goal of this survey is to provide an overarching understanding of how current approaches are tackling the resource challenges posed by large foundation models and to potentially inspire future breakthroughs in this field.

Li Zhang, Hainiu Xu, Abhinav Kommula et al.

Much text describes a changing world (e.g., procedures, stories, newswires), and understanding them requires tracking how entities change. An earlier dataset, OpenPI, provided crowdsourced annotations of entity state changes in text. However, a major limitation was that those annotations were free-form and did not identify salient changes, hampering model evaluation. To overcome these limitations, we present an improved dataset, OpenPI2.0, where entities and attributes are fully canonicalized and additional entity salience annotations are added. On our fairer evaluation setting, we find that current state-of-the-art language models are far from competent. We also show that using state changes of salient entities as a chain-of-thought prompt, downstream performance is improved on tasks such as question answering and classical planning, outperforming the setting involving all related entities indiscriminately. We offer OpenPI2.0 for the continued development of models that can understand the dynamics of entities in text.

Xiangyang Mou, Mo Yu, Shiyu Chang et al.

This paper proposes a new problem of complementary evidence identification for open-domain question answering (QA). The problem aims to efficiently find a small set of passages that covers full evidence from multiple aspects as to answer a complex question. To this end, we proposes a method that learns vector representations of passages and models the sufficiency and diversity within the selected set, in addition to the relevance between the question and passages. Our experiments demonstrate that our method considers the dependence within the supporting evidence and significantly improves the accuracy of complementary evidence selection in QA domain.

Li Zhang, Qing Lyu, Chris Callison-Burch

Modern task-oriented dialog systems need to reliably understand users' intents. Intent detection is most challenging when moving to new domains or new languages, since there is little annotated data. To address this challenge, we present a suite of pretrained intent detection models. Our models are able to predict a broad range of intended goals from many actions because they are trained on wikiHow, a comprehensive instructional website. Our models achieve state-of-the-art results on the Snips dataset, the Schema-Guided Dialogue dataset, and all 3 languages of the Facebook multilingual dialog datasets. Our models also demonstrate strong zero- and few-shot performance, reaching over 75% accuracy using only 100 training examples in all datasets.

Li Zhang, Han Wang, Lingxiao Li

Pair-based metric learning has been widely adopted to learn sentence embedding in many NLP tasks such as semantic text similarity due to its efficiency in computation. Most existing works employed a sequence encoder model and utilized limited sentence pairs with a pair-based loss to learn discriminating sentence representation. However, it is known that the sentence representation can be biased when the sampled sentence pairs deviate from the true distribution of all sentence pairs. In this paper, our theoretical analysis shows that existing works severely suffered from a good pair sampling and instance weighting strategy. Instead of one time pair selection and learning on equal weighted pairs, we propose a unified locality weighting and learning framework to learn task-specific sentence embedding. Our model, SentPWNet, exploits the neighboring spatial distribution of each sentence as locality weight to indicate the informative level of sentence pair. Such weight is updated along with pair-loss optimization in each round, ensuring the model keep learning the most informative sentence pairs. Extensive experiments on four public available datasets and a self-collected place search benchmark with 1.4 million places clearly demonstrate that our model consistently outperforms existing sentence embedding methods with comparable efficiency.

John Anderson, Qingqing Huang, Walid Krichene et al.

We extend the idea of word pieces in natural language models to machine learning tasks on opaque ids. This is achieved by applying hash functions to map each id to multiple hash tokens in a much smaller space, similarly to a Bloom filter. We show that by applying a multi-layer Transformer to these Bloom filter digests, we are able to obtain models with high accuracy. They outperform models of a similar size without hashing and, to a large degree, models of a much larger size trained using sampled softmax with the same computational budget. Our key observation is that it is important to use a multi-layer Transformer for Bloom filter digests to remove ambiguity in the hashed input. We believe this provides an alternative method to solving problems with large vocabulary size.

Walid Krichene, Nicolas Mayoraz, Steffen Rendle et al.

We study the problem of learning similarity functions over very large corpora using neural network embedding models. These models are typically trained using SGD with sampling of random observed and unobserved pairs, with a number of samples that grows quadratically with the corpus size, making it expensive to scale to very large corpora. We propose new efficient methods to train these models without having to sample unobserved pairs. Inspired by matrix factorization, our approach relies on adding a global quadratic penalty to all pairs of examples and expressing this term as the matrix-inner-product of two generalized Gramians. We show that the gradient of this term can be efficiently computed by maintaining estimates of the Gramians, and develop variance reduction schemes to improve the quality of the estimates. We conduct large-scale experiments that show a significant improvement in training time and generalization quality compared to traditional sampling methods.

Catherine Finegan-Dollak, Jonathan K. Kummerfeld, Li Zhang et al.

To be informative, an evaluation must measure how well systems generalize to realistic unseen data. We identify limitations of and propose improvements to current evaluations of text-to-SQL systems. First, we compare human-generated and automatically generated questions, characterizing properties of queries necessary for real-world applications. To facilitate evaluation on multiple datasets, we release standardized and improved versions of seven existing datasets and one new text-to-SQL dataset. Second, we show that the current division of data into training and test sets measures robustness to variations in the way questions are asked, but only partially tests how well systems generalize to new queries; therefore, we propose a complementary dataset split for evaluation of future work. Finally, we demonstrate how the common practice of anonymizing variables during evaluation removes an important challenge of the task. Our observations highlight key difficulties, and our methodology enables effective measurement of future development.

Li Zhang, Steven R. Wilson, Rada Mihalcea

Multi-relational semantic similarity datasets define the semantic relations between two short texts in multiple ways, e.g., similarity, relatedness, and so on. Yet, all the systems to date designed to capture such relations target one relation at a time. We propose a multi-label transfer learning approach based on LSTM to make predictions for several relations simultaneously and aggregate the losses to update the parameters. This multi-label regression approach jointly learns the information provided by the multiple relations, rather than treating them as separate tasks. Not only does this approach outperform the single-task approach and the traditional multi-task learning approach, but it also achieves state-of-the-art performance on all but one relation of the Human Activity Phrase dataset.

Li Zhang, Steven R. Wilson, Rada Mihalcea

Sentence encoders, which produce sentence embeddings using neural networks, are typically evaluated by how well they transfer to downstream tasks. This includes semantic similarity, an important task in natural language understanding. Although there has been much work dedicated to building sentence encoders, the accompanying transfer learning techniques have received relatively little attention. In this paper, we propose a transfer learning setting specialized for semantic similarity, which we refer to as direct network transfer. Through experiments on several standard text similarity datasets, we show that applying direct network transfer to existing encoders can lead to state-of-the-art performance. Additionally, we compare several approaches to transfer sentence encoders to semantic similarity tasks, showing that the choice of transfer learning setting greatly affects the performance in many cases, and differs by encoder and dataset.