Author name cluster

Jeevana Priya Inala

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

8 papers

2 author rows

ICLR Conference 2024 Conference Paper

Is Self-Repair a Silver Bullet for Code Generation?

Theo X. Olausson
Jeevana Priya Inala
Chenglong Wang 0005
Jianfeng Gao 0001
Armando Solar-Lezama

Large language models have shown remarkable aptitude in code generation, but still struggle to perform complex tasks. Self-repair---in which the model debugs and repairs its own code---has recently become a popular way to boost performance in these settings. However, despite its increasing popularity, existing studies of self-repair have been limited in scope; in many settings, its efficacy thus remains poorly understood. In this paper, we analyze Code Llama, GPT-3.5 and GPT-4's ability to perform self-repair on problems taken from HumanEval and APPS. We find that when the cost of carrying out repair is taken into account, performance gains are often modest, vary a lot between subsets of the data, and are sometimes not present at all. We hypothesize that this is because self-repair is bottlenecked by the model's ability to provide feedback on its own code; using a stronger model to artificially boost the quality of the feedback, we observe substantially larger performance gains. Similarly, a small-scale study in which we provide GPT-4 with feedback from human participants suggests that even for the strongest models, self-repair still lags far behind what can be achieved with human-level debugging.

Details

ICLR Conference 2023 Conference Paper

Learning Math Reasoning from Self-Sampled Correct and Partially-Correct Solutions

Ansong Ni
Jeevana Priya Inala
Chenglong Wang 0005
Alex Polozov
Christopher Meek
Dragomir Radev
Jianfeng Gao 0001

Pretrained language models have shown superior performance on many natural language processing tasks, yet they still struggle at multi-step formal reasoning tasks like grade school math problems. One key challenge of finetuning them to solve such math reasoning problems is that many existing datasets only contain one reference solution for each problem, despite the fact that there are often alternative solutions resembling different reasoning paths to the final answer. This way, the finetuned models are biased towards the limited reference solutions, which limits their generalization to unseen examples. To mitigate this issue, we propose to let the model perform sampling during training and learn from both self-sampled fully-correct solutions, which yield the correct answer upon execution, and partially-correct solutions, whose intermediate state matches an intermediate state of a known correct solution. We show that our use of self-sampled correct and partially-correct solutions can benefit learning and help guide the sampling process, leading to more efficient exploration of the solution space. Additionally, we explore various training objectives to support learning from multiple solutions per example and find they greatly affect the performance. Experiments on two math reasoning datasets show the effectiveness of our method compared to learning from a single reference solution with MLE, where we improve PASS@100 from 35.5% to 44.5% for GSM8K, and 27.6% to 36.2% PASS@80 for MathQA. Such improvements are also consistent across different model sizes.

Details

NeurIPS Conference 2022 Conference Paper

Fault-Aware Neural Code Rankers

Jeevana Priya Inala
Chenglong Wang
Mei Yang
Andres Codas
Mark Encarnación
Shuvendu Lahiri
Madanlal Musuvathi
Jianfeng Gao

Large language models (LLMs) have demonstrated an impressive ability to generate code for various programming tasks. In many instances, LLMs can generate a correct program for a task when given numerous trials. Consequently, a recent trend is to do large scale sampling of programs using a model and then filtering/ranking the programs based on the program execution on a small number of known unit tests to select one candidate solution. However, these approaches assume that the unit tests are given and assume the ability to safely execute the generated programs (which can do arbitrary dangerous operations such as file manipulations). Both of the above assumptions are impractical in real-world software development. In this paper, we propose CodeRanker, a neural ranker that can predict the correctness of a sampled program without executing it. Our CodeRanker is fault-aware i. e. , it is trained to predict different kinds of execution information such as predicting the exact compile/runtime error type (e. g. , an IndexError or a TypeError). We show that CodeRanker can significantly increase the pass@1 accuracy of various code generation models (including Codex, GPT-Neo, GPT-J) on APPS, HumanEval and MBPP datasets.

PDF Details

NeurIPS Conference 2021 Conference Paper

Program Synthesis Guided Reinforcement Learning for Partially Observed Environments

Yichen Yang
Jeevana Priya Inala
Osbert Bastani
Yewen Pu
Armando Solar-Lezama
Martin Rinard

A key challenge for reinforcement learning is solving long-horizon planning problems. Recent work has leveraged programs to guide reinforcement learning in these settings. However, these approaches impose a high manual burden on the user since they must provide a guiding program for every new task. Partially observed environments further complicate the programming task because the program must implement a strategy that correctly, and ideally optimally, handles every possible configuration of the hidden regions of the environment. We propose a new approach, model predictive program synthesis (MPPS), that uses program synthesis to automatically generate the guiding programs. It trains a generative model to predict the unobserved portions of the world, and then synthesizes a program based on samples from this model in a way that is robust to its uncertainty. In our experiments, we show that our approach significantly outperforms non-program-guided approaches on a set of challenging benchmarks, including a 2D Minecraft-inspired environment where the agent must complete a complex sequence of subtasks to achieve its goal, and achieves a similar performance as using handcrafted programs to guide the agent. Our results demonstrate that our approach can obtain the benefits of program-guided reinforcement learning without requiring the user to provide a new guiding program for every new task.

PDF Details

NeurIPS Conference 2020 Conference Paper

Neurosymbolic Transformers for Multi-Agent Communication

Jeevana Priya Inala
Yichen Yang
James Paulos
Yewen Pu
Osbert Bastani
Vijay Kumar
Martin Rinard
Armando Solar-Lezama

We study the problem of inferring communication structures that can solve cooperative multi-agent planning problems while minimizing the amount of communication. We quantify the amount of communication as the maximum degree of the communication graph; this metric captures settings where agents have limited bandwidth. Minimizing communication is challenging due to the combinatorial nature of both the decision space and the objective; for instance, we cannot solve this problem by training neural networks using gradient descent. We propose a novel algorithm that synthesizes a control policy that combines a programmatic communication policy used to generate the communication graph with a transformer policy network used to choose actions. Our algorithm first trains the transformer policy, which implicitly generates a "soft" communication graph; then, it synthesizes a programmatic communication policy that "hardens" this graph, forming a neurosymbolic transformer. Our experiments demonstrate how our approach can synthesize policies that generate low-degree communication graphs while maintaining near-optimal performance.

PDF Details

ICLR Conference 2020 Conference Paper

Synthesizing Programmatic Policies that Inductively Generalize

Jeevana Priya Inala
Osbert Bastani
Zenna Tavares
Armando Solar-Lezama

Deep reinforcement learning has successfully solved a number of challenging control tasks. However, learned policies typically have difficulty generalizing to novel environments. We propose an algorithm for learning programmatic state machine policies that can capture repeating behaviors. By doing so, they have the ability to generalize to instances requiring an arbitrary number of repetitions, a property we call inductive generalization. However, state machine policies are hard to learn since they consist of a combination of continuous and discrete structures. We propose a learning framework called adaptive teaching, which learns a state machine policy by imitating a teacher; in contrast to traditional imitation learning, our teacher adaptively updates itself based on the structure of the student. We show that our algorithm can be used to learn policies that inductively generalize to novel environments, whereas traditional neural network policies fail to do so.

Details

ICRA Conference 2019 Conference Paper

Task-Based Design of Ad-hoc Modular Manipulators

Thais Campos
Jeevana Priya Inala
Armando Solar-Lezama
Hadas Kress-Gazit

The great promise of modular robots is the ability to create on demand robots; however, choosing the “right” design based on a task is still a challenging problem. In this paper, we present an approach to automatically synthesize both the design and control for modular robots from a task description. In particular, we focus on manipulators composed of one degree-of-freedom (DoF) modules. Our approach is able to handle partially infeasible tasks by either identifying the infeasible part and finding a design that satisfies the feasible part or searching for multiple designs that together satisfy the entire task. We compare our approach to a baseline genetic algorithm in a series of increasingly complex environments.

Details

SAT Conference 2016 Conference Paper

Synthesis of Domain Specific CNF Encoders for Bit-Vector Solvers

Jeevana Priya Inala
Rohit Singh 0002
Armando Solar-Lezama

Abstract The theory of bit-vectors in SMT solvers is very important for many applications due to its ability to faithfully model the behavior of machine instructions. A crucial step in solving bit-vector formulas is the translation from high-level bit-vector terms down to low-level boolean formulas that can be efficiently mapped to CNF clauses and fed into a SAT solver. In this paper, we demonstrate how a combination of program synthesis and machine learning technology can be used to automatically generate code to perform this translation in a way that is tailored to particular problem domains. Using this technique, the paper shows that we can improve upon the basic encoding strategy used by CVC4 (a state of the art SMT solver) and automatically generate variants of the solver tailored to different domains of problems represented in the bit-vector benchmark suite from SMT-COMP 2015.

Details