MATHSENSEI: A Tool-Augmented Large Language Model for Mathematical Reasoning

Tool-augmented Large Language Models (TALM) are known to enhance the skillset of large language models (LLM), thereby, leading to their improved reasoning abilities across many tasks. While, TALMs have been successfully employed in different question-answering benchmarks, their efficacy on complex mathematical reasoning benchmarks, and the potential complimentary benefits offered by tools for knowledge retrieval and mathematical equation solving, are open research questions. In this work, we present MATHSENSEI, a tool-augmented large language model for mathematical reasoning. Augmented with tools for knowledge retrieval (Bing Web Search), program execution (Python), and symbolic equation solving (Wolfram-Alpha), we study the complimentary benefits of these tools through evaluations on mathematical reasoning datasets. We perform exhaustive ablations on MATH,a popular dataset for evaluating mathematical reasoning on diverse mathematical disciplines. We also conduct experiments involving well-known tool planners to study the impact of tool sequencing on the model performance. MATHSENSEI achieves 13.5% better accuracy over gpt-3.5-turbo with chain-of-thought on the MATH dataset. We further observe that TALMs are not as effective for simpler math word problems (in GSM-8k), and the benefit increases as the complexity and required knowledge increases (progressively over AQuA, MMLU-Math, and higher level complex questions in MATH). The code and data are available at https://github.com/Debrup-61/MathSensei.

GRAFFORD: A Benchmark Dataset for Testing the Knowledge of Object Affordances of Language and Vision Models

We investigate the knowledge of object affordances in pre-trained language models (LMs) and pre-trained Vision-Language models (VLMs). Transformers-based large pre-trained language models (PTLM) learn contextual representation from massive amounts of unlabeled text and are shown to perform impressively in downstream NLU tasks. In parallel, a growing body of literature shows that PTLMs fail inconsistently and non-intuitively, showing a lack of reasoning and grounding. To take a first step toward quantifying the effect of grounding (or lack thereof), we curate a novel and comprehensive dataset of object affordances -- GrAFFORD, characterized by 15 affordance classes. Unlike affordance datasets collected in vision and language domains, we annotate in-the-wild sentences with objects and affordances. Experimental results reveal that PTLMs exhibit limited reasoning abilities when it comes to uncommon object affordances. We also observe that pre-trained VLMs do not necessarily capture object affordances effectively. Through few-shot fine-tuning, we demonstrate improvement in affordance knowledge in PTLMs and VLMs. Our research contributes a novel dataset for language grounding tasks, and presents insights into LM capabilities, advancing the understanding of object affordances. Codes and data are available at https://github.com/sayantan11995/Affordance

Code Prompting Elicits Conditional Reasoning Abilities in Text+Code LLMs

Reasoning is a fundamental component for achieving language understanding. Among the multiple types of reasoning, conditional reasoning, the ability to draw different conclusions depending on some condition, has been understudied in large language models (LLMs). Recent prompting methods, such as chain of thought, have significantly improved LLMs on reasoning tasks. Nevertheless, there is still little understanding of what triggers reasoning abilities in LLMs. We hypothesize that code prompts can trigger conditional reasoning in LLMs trained on text and code. We propose a chain of prompts that transforms a natural language problem into code and prompts the LLM with the generated code. Our experiments find that code prompts exhibit a performance boost between 2.6 and 7.7 points on GPT 3.5 across multiple datasets requiring conditional reasoning. We then conduct experiments to discover how code prompts elicit conditional reasoning abilities and through which features. We observe that prompts need to contain natural language text accompanied by high-quality code that closely represents the semantics of the instance text. Furthermore, we show that code prompts are more efficient, requiring fewer demonstrations, and that they trigger superior state tracking of variables or key entities.

Stuck in the Quicksand of Numeracy, Far from AGI Summit: Evaluating LLMs' Mathematical Competency through Ontology-guided Perturbations

Recent advancements in Large Language Models (LLMs) have showcased striking results on existing logical reasoning benchmarks, with some models even surpassing human performance. However, the true depth of their competencies and robustness, in mathematical reasoning tasks, remains an open question. In response, we develop (i) an ontology of perturbations of maths questions, (ii) a semi-automatic method of perturbation, and (iii) a dataset of perturbed maths questions to probe the limits of LLM capabilities in mathematical reasoning tasks. These controlled perturbations span across multiple fine dimensions of the structural and representational aspects of maths questions. Using GPT-4, we generated the MORE dataset by perturbing randomly selected five seed questions from GSM8K. This process was guided by our ontology and involved a thorough automatic and manual filtering process, yielding a set of 216 maths problems. We conducted comprehensive evaluation of both closed-source and open-source LLMs on MORE. The results show a significant performance drop across all the models against the perturbed questions. This strongly suggests that current LLMs lack robust mathematical skills and deep reasoning abilities. This research not only identifies multiple gaps in the capabilities of current models, but also highlights multiple potential directions for future development. Our dataset will be made publicly available at https://huggingface.co/datasets/declare-lab/GSM8k_MORE.

Towards LogiGLUE: A Brief Survey and A Benchmark for Analyzing Logical Reasoning Capabilities of Language Models

Logical reasoning is fundamental for humans yet presents a substantial challenge in the domain of Artificial Intelligence. Initially, researchers used Knowledge Representation and Reasoning (KR) systems that did not scale and required non trivial manual effort. Recently, the emergence of large language models (LLMs) has demonstrated the ability to overcome various limitations of formal Knowledge Representation (KR) systems. Consequently, there is a growing interest in using LLMs for logical reasoning via natural language. This work strives to understand the proficiency of LLMs in logical reasoning by offering a brief review of the latest progress in this area; with a focus on the logical reasoning datasets, tasks, and the methods adopted to utilize LLMs for reasoning. To offer a thorough analysis, we have compiled a benchmark titled LogiGLUE. This includes 24 varied datasets encompassing deductive, abductive, and inductive reasoning. We have standardized these datasets into Seq2Seq tasks to facilitate straightforward training and evaluation for future research. Utilizing LogiGLUE as a foundation, we have trained an instruction fine tuned language model, resulting in LogiT5. We study single task training, multi task training, and a chain of thought knowledge distillation fine tuning technique to assess the performance of model across the different logical reasoning categories. By this comprehensive process, we aim to shed light on the capabilities and potential pathways for enhancing logical reasoning proficiency in LLMs, paving the way for more advanced and nuanced developments in this critical field.

Tricking LLMs into Disobedience: Understanding, Analyzing, and Preventing Jailbreaks

Recent explorations with commercial Large Language Models (LLMs) have shown that non-expert users can jailbreak LLMs by simply manipulating the prompts; resulting in degenerate output behavior, privacy and security breaches, offensive outputs, and violations of content regulator policies. Limited formal studies have been carried out to formalize and analyze these attacks and their mitigations. We bridge this gap by proposing a formalism and a taxonomy of known (and possible) jailbreaks. We perform a survey of existing jailbreak methods and their effectiveness on open-source and commercial LLMs (such as GPT 3.5, OPT, BLOOM, and FLAN-T5-xxl). We further propose a limited set of prompt guards and discuss their effectiveness against known attack types.

ReMask: A Robust Information-Masking Approach for Domain Counterfactual Generation

Domain shift is a big challenge in NLP, thus, many approaches resort to learning domain-invariant features to mitigate the inference phase domain shift. Such methods, however, fail to leverage the domain-specific nuances relevant to the task at hand. To avoid such drawbacks, domain counterfactual generation aims to transform a text from the source domain to a given target domain. However, due to the limited availability of data, such frequency-based methods often miss and lead to some valid and spurious domain-token associations. Hence, we employ a three-step domain obfuscation approach that involves frequency and attention norm-based masking, to mask domain-specific cues, and unmasking to regain the domain generic context. Our experiments empirically show that the counterfactual samples sourced from our masked text lead to improved domain transfer on 10 out of 12 domain sentiment classification settings, with an average of 2% accuracy improvement over the state-of-the-art for unsupervised domain adaptation (UDA). Further, our model outperforms the state-of-the-art by achieving 1.4% average accuracy improvement in the adversarial domain adaptation (ADA) setting. Moreover, our model also shows its domain adaptation efficacy on a large multi-domain intent classification dataset where it attains state-of-the-art results. We release the codes publicly at \url{https://github.com/declare-lab/remask}.

Generating Intermediate Steps for NLI with Next-Step Supervision

The Natural Language Inference (NLI) task often requires reasoning over multiple steps to reach the conclusion. While the necessity of generating such intermediate steps (instead of a summary explanation) has gained popular support, it is unclear how to generate such steps without complete end-to-end supervision and how such generated steps can be further utilized. In this work, we train a sequence-to-sequence model to generate only the next step given an NLI premise and hypothesis pair (and previous steps); then enhance it with external knowledge and symbolic search to generate intermediate steps with only next-step supervision. We show the correctness of such generated steps through automated and human verification. Furthermore, we show that such generated steps can help improve end-to-end NLI task performance using simple data augmentation strategies, across multiple public NLI datasets.

Multilingual CheckList: Generation and Evaluation

The recently proposed CheckList (Riberio et al,. 2020) approach to evaluation of NLP systems has revealed high failure rates for basic capabilities for multiple state-of-the-art and commercial models. However, the CheckList creation process is manual which creates a bottleneck towards creation of multilingual CheckLists catering 100s of languages. In this work, we explore multiple approaches to generate and evaluate the quality of Multilingual CheckList. We device an algorithm -- Automated Multilingual Checklist Generation (AMCG) for automatically transferring a CheckList from a source to a target language that relies on a reasonable machine translation system. We then compare the CheckList generated by AMCG with CheckLists generated with different levels of human intervention. Through in-depth crosslingual experiments between English and Hindi, and broad multilingual experiments spanning 11 languages, we show that the automatic approach can provide accurate estimates of failure rates of a model across capabilities, as would a human-verified CheckList, and better than CheckLists generated by humans from scratch.

LoNLI: An Extensible Framework for Testing Diverse Logical Reasoning Capabilities for NLI

Natural Language Inference (NLI) is considered a representative task to test natural language understanding (NLU). In this work, we propose an extensible framework to collectively yet categorically test diverse Logical reasoning capabilities required for NLI (and by extension, NLU). Motivated by behavioral testing, we create a semi-synthetic large test-bench (363 templates, 363k examples) and an associated framework that offers following utilities: 1) individually test and analyze reasoning capabilities along 17 reasoning dimensions (including pragmatic reasoning), 2) design experiments to study cross-capability information content (leave one out or bring one in); and 3) the synthetic nature enable us to control for artifacts and biases. The inherited power of automated test case instantiation from free-form natural language templates (using CheckList), and a well-defined taxonomy of capabilities enable us to extend to (cognitively) harder test cases while varying the complexity of natural language. Through our analysis of state-of-the-art NLI systems, we observe that our benchmark is indeed hard (and non-trivial even with training on additional resources). Some capabilities stand out as harder. Further fine-grained analysis and fine-tuning experiments reveal more insights about these capabilities and the models -- supporting and extending previous observations. Towards the end we also perform an user-study, to investigate whether behavioral information can be utilised to generalize much better for some models compared to others.