Author name cluster

Sriram Rajamani

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.

4 papers

1 author row

NeurIPS Conference 2023 Conference Paper

Monitor-Guided Decoding of Code LMs with Static Analysis of Repository Context

Lakshya A Agrawal
Aditya Kanade
Navin Goyal
Shuvendu Lahiri
Sriram Rajamani

Language models of code (LMs) work well when the surrounding code provides sufficient context. This is not true when it becomes necessary to use types, functionality or APIs defined elsewhere in the repository or a linked library, especially those not seen during training. LMs suffer from limited awareness of such global context and end up hallucinating. Integrated development environments (IDEs) assist developers in understanding repository context using static analysis. We extend this assistance, enjoyed by developers, to LMs. We propose monitor-guided decoding (MGD) where a monitor uses static analysis to guide the decoding. We construct a repository-level dataset PragmaticCode for method-completion in Java and evaluate MGD on it. On models of varying parameter scale, by monitoring for type-consistent object dereferences, MGD consistently improves compilation rates and agreement with ground truth. Further, LMs with fewer parameters, when augmented with MGD, can outperform larger LMs. With MGD, SantaCoder-1. 1B achieves better compilation rate and next-identifier match than the much larger text-davinci-003 model. We also conduct a generalizability study to evaluate the ability of MGD to generalize to multiple programming languages (Java, C# and Rust), coding scenarios (e. g. , correct number of arguments to method calls), and to enforce richer semantic constraints (e. g. , stateful API protocols). Our data and implementation are available at https: //github. com/microsoft/monitors4codegen.

PDF Details

AAAI Conference 2014 Conference Paper

R2: An Efficient MCMC Sampler for Probabilistic Programs

Aditya Nori
Chung-Kil Hur
Sriram Rajamani
Selva Samuel

We present a new Markov Chain Monte Carlo (MCMC) sampling algorithm for probabilistic programs. Our approach and tool, called R2, has the unique feature of employing program analysis in order to improve the efficiency of MCMC sampling. Given an input program P, R2 propagates observations in P backwards to obtain a semantically equivalent program P0 in which every probabilistic assignment is immediately followed by an observe statement. Inference is performed by a suitably modified version of the Metropolis-Hastings algorithm that exploits the structure of the program P0. This has the overall effect of preventing rejections due to program executions that fail to satisfy observations in P. We formalize the semantics of probabilistic programs and rigorously prove the correctness of R2. We also empirically demonstrate the effectiveness of R2—in particular, we show that R2 is able to produce results of similar quality as the CHURCH and STAN probabilistic programming tools with much shorter execution time.

PDF Details

AAAI Conference 2012 Conference Paper

Automatically Generating Algebra Problems

Rohit Singh
Sumit Gulwani
Sriram Rajamani

We propose computer-assisted techniques for helping with pedagogy in Algebra. In particular, given a proof problem p (of the form Left-hand-side-term = Righthand-side-term), we show how to automatically generate problems that are similar to p. We believe that such a tool can be used by teachers in making examinations where they need to test students on problems similar to what they taught in class, and by students in generating practice problems tailored to their specific needs. Our first insight is that we can generalize p syntactically to a query Q that implicitly represents a set of problems [[Q]] (which includes p). Our second insight is that we can explore the space of problems [[Q]] automatically, use classical results from polynomial identity testing to generate only those problems in [[Q]] that are correct, and then use pruning techniques to generate only unique and interesting problems. Our third insight is that with a small amount of manual tuning on the query Q, the user can interactively guide the computer to generate problems of interest to her. We present the technical details of the above mentioned steps, and also describe a tool where these steps have been implemented. We also present an empirical evaluation on a wide variety of problems from various sub-fields of algebra including polynomials, trigonometry, calculus, determinants etc. Our tool is able to generate a rich corpus of similar problems from each given problem; while some of these similar problems were already present in the textbook, several were new!

PDF Details

IS Journal 2006 Journal Article

Intelligent Systems and Formal Methods in Software Engineering

Bernhard Beckert
Tony Hoare
Reiner Hahnle
Douglas Smith
Cordell Green
Silvio Ranise
Cesare Tinelli
Thomas Ball

Over the last few years, technologies for the formal description, construction, analysis, and validation of software - based mostly on logics and formal reasoning - have matured. We can expect them to complement and partly replace traditional software engineering methods in the future. Formal methods in software engineering are an increasingly important application area for intelligent systems. The field has outgrown the area of academic case studies, and industry is showing serious interest. We convincingly argue that we've reached the point where we can solve the problem of how to formally verify industrial-scale software. We propose program verification as a computer science Grand Challenge. Deductive software verification is a core technology of formal methods. We describe recent dramatic changes in the way it's perceived and used. Another important base technique of formal methods, besides software verification, is synthesizing software that's correct by construction because it's formally derived from its specification. We discuss recent developments and trends in this area. Surprisingly efficient decision procedures for the satisfiability modulo theories problem have recently emerged. We explain these techniques and why they're important for all formal-methods tools. We look at formal methods from an industry perspective. We explain the success of Microsoft Research's SLAM project, which has developed a verification tool for device drivers

Details DOI