SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.07381 · 2022

SeqZero: Few-shot Compositional Semantic Parsing with Sequential Prompts and Zero-shot Models

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2205.07381.json · badge: /api/ran/2205.07381.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
amzn/seqzero canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
build_partial_prefixes_from_file Ran
ran; our contract draft was wrong, not the code
amzn/seqzero/parser/semantic_parsing.py
code served (permissive licence) · get_code("a683bb102e3ba6f7")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned list. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import tempfile
import os

torch.manual_seed(0)

class MockDictionary:
    def encode_line(self, line, append_eos=False):
        tokens = [hash(word) % 1000 for word in line.split()]
        return torch.tensor(tokens, dtype=torch.long)

class MockBPE:
    def encode(self, text):
        words = text.split()
        return ' '.join([f'bpe_{i}' for i in range(len(words))])

class MockTask:
    def __init__(self):
        self.source_dictionary = MockDictionary()

class MockBART(nn.Module):
    def __init__(self):
        super().__init__()
        self.bpe = MockBPE()
        self.task = MockTask()
        self.max_positions = [100, 100]

bart = MockBART()

temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt', encoding='utf-8')
temp_file.write("hello world\n")
temp_file.write("this is a test\n")
temp_file.write("short\n")
temp_file.close()

file_path = temp_file.name

ARGS = {
    'bart': bart,
    'file': file_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
build_prefixes_from_file Ran
ran, honoured the contract we drafted
amzn/seqzero/parser/semantic_parsing.py
code served (permissive licence) · get_code("a686be534935090b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned list, ndim 2, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from collections import namedtuple
import tempfile
import os

torch.manual_seed(0)

# Create a minimal BPE encoder stand-in
class MinimalBPE:
    def encode(self, text):
        # Return a simple space-separated string of tokens
        tokens = text.split()
        return ' '.join([f'tok_{i}' for i in range(len(tokens))])

# Create a minimal dictionary for encoding
class MinimalDictionary:
    def __init__(self):
        self.indices = {}
        self.symbols = []
        # Pre-populate with special tokens and some common tokens
        for token in ['</s>', '<s>', 'tok_0', 'tok_1', 'tok_2', 'tok_3', 'tok_4']:
            if token not in self.indices:
                self.indices[token] = len(self.symbols)
                self.symbols.append(token)
    
    def encode_line(self, line, append_eos=False):
        tokens = line.split()
        indices = []
        for token in tokens:
            if token not in self.indices:
                self.indices[token] = len(self.symbols)
                self.symbols.append(token)
            indices.append(self.indices[token])
        result = torch.tensor(indices, dtype=torch.long)
        return result

# Create a minimal task
class MinimalTask:
    def __init__(self):
        self.source_dictionary = MinimalDictionary()

# Create a minimal BART model
class MinimalBART(nn.Module):
    def __init__(self):
        super().__init__()
        self.bpe = MinimalBPE()
        self.task = MinimalTask()
        self.max_positions = [1024, 1024]  # min(max_positions) = 1024
    
    def forward(self, x):
        return x

# Create a temporary file with test data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt', encoding='utf-8')
temp_file.write("hello world test\n")
temp_file.write("another test sentence\n")
temp_file.close()

bart = MinimalBART()
file = temp_file.name

ARGS = {
    'bart': bart,
    'file': file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 2 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recent research showed promising results on combining pretrained language models (LMs) with canonical utterance for few-shot semantic parsing. The canonical utterance is often lengthy and complex due to the compositional structure of formal languages. Learning to generate such canonical utterance requires significant amount of data to reach high performance. Fine-tuning with only few-shot samples, the LMs can easily forget pretrained knowledge, overfit spurious biases, and suffer from compositionally out-of-distribution generalization errors. To tackle these issues, we propose a novel few-shot semantic parsing method -- SeqZero. SeqZero decomposes the problem into a sequence of sub-problems, which correspond to the sub-clauses of the formal language. Based on the decomposition, the LMs only need to generate short answers using prompts for predicting sub-clauses. Thus, SeqZero avoids generating a long canonical utterance at once. Moreover, SeqZero employs not only a few-shot model but also a zero-shot model to alleviate the overfitting. In particular, SeqZero brings out the merits from both models via ensemble equipped with our proposed constrained rescaling. SeqZero achieves SOTA performance of BART-based models on GeoQuery and EcommerceQuery, which are two few-shot datasets with compositional data split.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2205.07381")
get_code_for_paper("2205.07381")
have("2205.07381")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2205.07381.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2205.07381.svg)](https://syntology.ai/paper/2205.07381)

Connect an agent — have() is free.