MASAI: Modular Architecture for Software-engineering AI Agents

A common method to solve complex problems in software engineering, is to divide the problem into multiple sub-problems. Inspired by this, we propose a Modular Architecture for Software-engineering AI (MASAI) agents, where different LLM-powered sub-agents are instantiated with well-defined objectives and strategies tuned to achieve those objectives. Our modular architecture offers several advantages: (1) employing and tuning different problem-solving strategies across sub-agents, (2) enabling sub-agents to gather information from different sources scattered throughout a repository, and (3) avoiding unnecessarily long trajectories which inflate costs and add extraneous context. MASAI enabled us to achieve the highest performance (28.33% resolution rate) on the popular and highly challenging SWE-bench Lite dataset consisting of 300 GitHub issues from 11 Python repositories. We conduct a comprehensive evaluation of MASAI relative to other agentic methods and analyze the effects of our design decisions and their contribution to the success of MASAI.

GAR-meets-RAG Paradigm for Zero-Shot Information Retrieval

Given a query and a document corpus, the information retrieval (IR) task is to output a ranked list of relevant documents. Combining large language models (LLMs) with embedding-based retrieval models, recent work shows promising results on the zero-shot retrieval problem, i.e., no access to labeled data from the target domain. Two such popular paradigms are generation-augmented retrieval or GAR (generate additional context for the query and then retrieve), and retrieval-augmented generation or RAG (retrieve relevant documents as context and then generate answers). The success of these paradigms hinges on (i) high-recall retrieval models, which are difficult to obtain in the zero-shot setting, and (ii) high-precision (re-)ranking models which typically need a good initialization. In this work, we propose a novel GAR-meets-RAG recurrence formulation that overcomes the challenges of existing paradigms. Our method iteratively improves retrieval (via GAR) and rewrite (via RAG) stages in the zero-shot setting. A key design principle is that the rewrite-retrieval stages improve the recall of the system and a final re-ranking stage improves the precision. We conduct extensive experiments on zero-shot passage retrieval benchmarks, BEIR and TREC-DL. Our method establishes a new state-of-the-art in the BEIR benchmark, outperforming previous best results in Recall@100 and nDCG@10 metrics on 6 out of 8 datasets, with up to 17% relative gains over the previous best.

Learning and Leveraging Verifiers to Improve Planning Capabilities of Pre-trained Language Models

There have been wide spread claims in the literature about the emergent reasoning capabilities of Pretrained Large Language Models. However, recent studies, have found that their ability to plan remains questionable. Through our experiments using GPT-2, we empirically demonstrate that the performance of a finetuned baseline remains poor because it violates pre-conditions of actions in the plans that it generates. To improve the planning capabilities of a finetuned LLM, we train a verifier, which can classify actions as being valid or invalid in a particular state. By randomly sampling actions from the same dataset, we generate examples of invalid actions which are then used to train a verifier which can check for action applicability. In the presence of diverse sampling from a generator and a verifier which can prune invalid trajectories, we show significant gains in the success rate on the Blocksworld domain. Additionally, we show that finetuning the GPT-2 generator itself to create the verifier generalizes better than finetuning the base GPT-2. Lastly, we investigate the role of the sampling temperature which can be used to control the exploration-exploitation tradeoff.

Have LLMs Advanced Enough? A Challenging Problem Solving Benchmark For Large Language Models

The performance on Large Language Models (LLMs) on existing reasoning benchmarks has shot up considerably over the past years. In response, we present JEEBench, a considerably more challenging benchmark dataset for evaluating the problem solving abilities of LLMs. We curate 450 challenging pre-engineering mathematics, physics and chemistry problems from the IIT JEE-Advanced exam. Long-horizon reasoning on top of deep in-domain knowledge is essential for solving problems in this benchmark. Our evaluation on the GPT series of models reveals that although performance improves with newer models, the best being GPT-4, the highest performance, even after using techniques like Self-Consistency and Chain-of-Thought prompting is less than 40 percent. Our analysis demonstrates that errors in algebraic manipulation and failure in retrieving relevant domain specific concepts are primary contributors to GPT4's low performance. Given the challenging nature of the benchmark, we hope that it can guide future research in problem solving using LLMs. Our code and dataset is available here.