Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More?

Long-context language models (LCLMs) have the potential to revolutionize our approach to tasks traditionally reliant on external tools like retrieval systems or databases. Leveraging LCLMs' ability to natively ingest and process entire corpora of information offers numerous advantages. It enhances user-friendliness by eliminating the need for specialized knowledge of tools, provides robust end-to-end modeling that minimizes cascading errors in complex pipelines, and allows for the application of sophisticated prompting techniques across the entire system. To assess this paradigm shift, we introduce LOFT, a benchmark of real-world tasks requiring context up to millions of tokens designed to evaluate LCLMs' performance on in-context retrieval and reasoning. Our findings reveal LCLMs' surprising ability to rival state-of-the-art retrieval and RAG systems, despite never having been explicitly trained for these tasks. However, LCLMs still face challenges in areas like compositional reasoning that are required in SQL-like tasks. Notably, prompting strategies significantly influence performance, emphasizing the need for continued research as context lengths grow. Overall, LOFT provides a rigorous testing ground for LCLMs, showcasing their potential to supplant existing paradigms and tackle novel tasks as model capabilities scale.

Every Answer Matters: Evaluating Commonsense with Probabilistic Measures

Large language models have demonstrated impressive performance on commonsense tasks; however, these tasks are often posed as multiple-choice questions, allowing models to exploit systematic biases. Commonsense is also inherently probabilistic with multiple correct answers. The purpose of "boiling water" could be making tea and cooking, but it also could be killing germs. Existing tasks do not capture the probabilistic nature of common sense. To this end, we present commonsense frame completion (CFC), a new generative task that evaluates common sense via multiple open-ended generations. We also propose a method of probabilistic evaluation that strongly correlates with human judgments. Humans drastically outperform strong language model baselines on our dataset, indicating this approach is both a challenging and useful evaluation of machine common sense.

Gecko: Versatile Text Embeddings Distilled from Large Language Models

We present Gecko, a compact and versatile text embedding model. Gecko achieves strong retrieval performance by leveraging a key idea: distilling knowledge from large language models (LLMs) into a retriever. Our two-step distillation process begins with generating diverse, synthetic paired data using an LLM. Next, we further refine the data quality by retrieving a set of candidate passages for each query, and relabeling the positive and hard negative passages using the same LLM. The effectiveness of our approach is demonstrated by the compactness of the Gecko. On the Massive Text Embedding Benchmark (MTEB), Gecko with 256 embedding dimensions outperforms all existing entries with 768 embedding size. Gecko with 768 embedding dimensions achieves an average score of 66.31, competing with 7x larger models and 5x higher dimensional embeddings.

Box Embeddings: An open-source library for representation learning using geometric structures

A major factor contributing to the success of modern representation learning is the ease of performing various vector operations. Recently, objects with geometric structures (eg. distributions, complex or hyperbolic vectors, or regions such as cones, disks, or boxes) have been explored for their alternative inductive biases and additional representational capacities. In this work, we introduce Box Embeddings, a Python library that enables researchers to easily apply and extend probabilistic box embeddings.

Word2Box: Learning Word Representation Using Box Embeddings

Learning vector representations for words is one of the most fundamental topics in NLP, capable of capturing syntactic and semantic relationships useful in a variety of downstream NLP tasks. Vector representations can be limiting, however, in that typical scoring such as dot product similarity intertwines position and magnitude of the vector in space. Exciting innovations in the space of representation learning have proposed alternative fundamental representations, such as distributions, hyperbolic vectors, or regions. Our model, Word2Box, takes a region-based approach to the problem of word representation, representing words as $n$-dimensional rectangles. These representations encode position and breadth independently and provide additional geometric operations such as intersection and containment which allow them to model co-occurrence patterns vectors struggle with. We demonstrate improved performance on various word similarity tasks, particularly on less common words, and perform a qualitative analysis exploring the additional unique expressivity provided by Word2Box.

Probabilistic Box Embeddings for Uncertain Knowledge Graph Reasoning

Knowledge bases often consist of facts which are harvested from a variety of sources, many of which are noisy and some of which conflict, resulting in a level of uncertainty for each triple. Knowledge bases are also often incomplete, prompting the use of embedding methods to generalize from known facts, however, existing embedding methods only model triple-level uncertainty, and reasoning results lack global consistency. To address these shortcomings, we propose BEUrRE, a novel uncertain knowledge graph embedding method with calibrated probabilistic semantics. BEUrRE models each entity as a box (i.e. axis-aligned hyperrectangle) and relations between two entities as affine transforms on the head and tail entity boxes. The geometry of the boxes allows for efficient calculation of intersections and volumes, endowing the model with calibrated probabilistic semantics and facilitating the incorporation of relational constraints. Extensive experiments on two benchmark datasets show that BEUrRE consistently outperforms baselines on confidence prediction and fact ranking due to its probabilistic calibration and ability to capture high-order dependencies among facts.

Modeling Fine-Grained Entity Types with Box Embeddings

Neural entity typing models typically represent entity types as vectors in a high-dimensional space, but such spaces are not well-suited to modeling these types' complex interdependencies. We study the ability of box embeddings, which represent entity types as d-dimensional hyperrectangles, to represent hierarchies of fine-grained entity type labels even when these relationships are not defined explicitly in the ontology. Our model represents both types and entity mentions as boxes. Each mention and its context are fed into a BERT-based model to embed that mention in our box space; essentially, this model leverages typological clues present in the surface text to hypothesize a type representation for the mention. Soft box containment can then be used to derive probabilities, both the posterior probability of a mention exhibiting a given type and the conditional probability relations between types themselves. We compare our approach with a strong vector-based typing model, and observe state-of-the-art performance on several entity typing benchmarks. In addition to competitive typing performance, our box-based model shows better performance in prediction consistency (predicting a supertype and a subtype together) and confidence (i.e., calibration), implying that the box-based model captures the latent type hierarchies better than the vector-based model does.

Improving Local Identifiability in Probabilistic Box Embeddings

Geometric embeddings have recently received attention for their natural ability to represent transitive asymmetric relations via containment. Box embeddings, where objects are represented by n-dimensional hyperrectangles, are a particularly promising example of such an embedding as they are closed under intersection and their volume can be calculated easily, allowing them to naturally represent calibrated probability distributions. The benefits of geometric embeddings also introduce a problem of local identifiability, however, where whole neighborhoods of parameters result in equivalent loss which impedes learning. Prior work addressed some of these issues by using an approximation to Gaussian convolution over the box parameters, however, this intersection operation also increases the sparsity of the gradient. In this work, we model the box parameters with min and max Gumbel distributions, which were chosen such that space is still closed under the operation of the intersection. The calculation of the expected intersection volume involves all parameters, and we demonstrate experimentally that this drastically improves the ability of such models to learn.

ProtoQA: A Question Answering Dataset for Prototypical Common-Sense Reasoning

Given questions regarding some prototypical situation -- such as Name something that people usually do before they leave the house for work? -- a human can easily answer them via acquired experiences. There can be multiple right answers for such questions with some more common for a situation than others. This paper introduces a new question answering dataset for training and evaluating common-sense reasoning capabilities of artificial intelligence systems in such prototypical situations. The training set is gathered from an existing set of questions played in a long-running international trivia game show -- Family Feud. The hidden evaluation set is created by gathering answers for each question from 100 crowd-workers. We also propose an open-domain task where a model has to output a ranked list of answers, ideally covering all prototypical answers for a question. On evaluating our dataset with various competitive state-of-the-art models, we find there is a significant gap between the best model and human performance on a number of evaluation metrics.

A Systematic Classification of Knowledge, Reasoning, and Context within the ARC Dataset

The recent work of Clark et al. introduces the AI2 Reasoning Challenge (ARC) and the associated ARC dataset that partitions open domain, complex science questions into an Easy Set and a Challenge Set. That paper includes an analysis of 100 questions with respect to the types of knowledge and reasoning required to answer them; however, it does not include clear definitions of these types, nor does it offer information about the quality of the labels. We propose a comprehensive set of definitions of knowledge and reasoning types necessary for answering the questions in the ARC dataset. Using ten annotators and a sophisticated annotation interface, we analyze the distribution of labels across the Challenge Set and statistics related to them. Additionally, we demonstrate that although naive information retrieval methods return sentences that are irrelevant to answering the query, sufficient supporting text is often present in the (ARC) corpus. Evaluating with human-selected relevant sentences improves the performance of a neural machine comprehension model by 42 points.