ASE 2023
Mon 11 - Fri 15 September 2023 Kirchberg, Luxembourg
Wed 13 Sep 2023 15:42 - 15:55 at Plenary Room 2 - Code Generation 2 Chair(s): Marianne Huchard

The performance of programming-by-example systems varies significantly across different tasks and even across different examples in one task. The key issue is that the search space depends on the given examples in a complex way. In particular, scalable synthesizers typically rely on a combination of machine learning to prioritize search order and deduction to prune search space, making it hard to quantitatively reason about how much an example speeds up the search. We propose a novel approach for quantifying the effectiveness of an example at reducing synthesis time. Based on this technique, we devise an algorithm that actively queries the user to obtain additional examples that significantly reduce synthesis time. We evaluate our approach on 30 challenging benchmarks across two different data science domains. Even with ineffective initial user-provided examples for pruning, our approach on average achieves a 6.0X speed-up in synthesis time compared to state-of-the-art synthesizers.

slides (faery-slides-v0.pdf)9.63MiB

Wed 13 Sep

Displayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change

15:30 - 17:00
15:30
12m
Talk
COMEX: A Tool for Generating Customized Source Code Representations
Tool Demonstrations
Debeshee Das Indian Institute of Technology Tirupati, Noble Saji Mathews University of Waterloo, Canada, Alex Mathai , Srikanth Tamilselvam IBM Research, Kranthi Sedamaki Indian Institute of Technology Tirupati, Sridhar Chimalakonda IIT Tirupati, Atul Kumar IBM India Research Labs
Pre-print Media Attached File Attached
15:42
12m
Talk
Fast and Reliable Program Synthesis via User Interaction
Research Papers
Yanju Chen University of California at Santa Barbara, Chenglong Wang Microsoft Research, Xinyu Wang University of Michigan, Osbert Bastani University of Pennsylvania, Yu Feng University of California at Santa Barbara
File Attached
15:55
12m
Talk
From Misuse to Mastery: Enhancing Code Generation with Knowledge-Driven AI Chaining
Research Papers
Xiaoxue Ren Zhejiang University, Xinyuan Ye Australian National University, Dehai Zhao CSIRO's Data61, Zhenchang Xing , Xiaohu Yang Zhejiang University
File Attached
16:08
12m
Talk
Generative Type Inference for PythonACM Distinguished Paper
Research Papers
Yun Peng Chinese University of Hong Kong, Chaozheng Wang The Chinese University of Hong Kong, Wenxuan Wang Chinese University of Hong Kong, Cuiyun Gao Harbin Institute of Technology, Michael Lyu The Chinese University of Hong Kong
Pre-print File Attached
16:21
12m
Talk
Compiler Auto-tuning via Critical Flag Selection
Research Papers
Mingxuan Zhu Peking University, Dan Hao Peking University
16:34
12m
Talk
Enhancing Code Safety in Quantum Intermediate Representation
NIER Track
Junjie Luo Kyushu University, Jianjun Zhao Kyushu University
File Attached
16:47
12m
Talk
CAT-LM: Training Language Models on Aligned Code And Tests
Research Papers
Nikitha Rao Carnegie Mellon University, Kush Jain Carnegie Mellon University, Uri Alon Carnegie Mellon University, Claire Le Goues Carnegie Mellon University, Vincent J. Hellendoorn Carnegie Mellon University
Media Attached File Attached