SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.15480 · ICLR · 2023

Post-hoc Concept Bottleneck Models

James Zou, Maggie Wang, Mert Yuksekgonul

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; 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/2205.15480.json · badge: /api/ran/2205.15480.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
mertyg/post-hoc-cbm community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
PosthocLinearCBM Ran
ran on a synthesized input · constructed an object; no result was computed
mertyg/post-hoc-cbm/models/pcbm_utils.py
code served (permissive licence) · get_code("0843a6f79624492a")
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, through the batch API) and returned an instance of its own class, PosthocLinearCBM: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

# Create a minimal ConceptBank stand-in class
class ConceptBank:
    def __init__(self, n_concepts=4, embedding_dim=3, n_classes=5):
        self.n_concepts = n_concepts
        self.embedding_dim = embedding_dim
        self.n_classes = n_classes
        
        # vectors: shape (n_concepts, embedding_dim)
        self.vectors = torch.randn(n_concepts, embedding_dim)
        
        # intercepts: shape (n_concepts,)
        self.intercepts = torch.randn(n_concepts)
        
        # norms: shape (n_concepts,)
        self.norms = torch.ones(n_concepts) * 2.0
        
        # concept_names: list of strings
        self.concept_names = [f"concept_{i}" for i in range(n_concepts)]

# Create concept bank instance
concept_bank = ConceptBank(n_concepts=4, embedding_dim=3, n_classes=5)

# Create idx_to_class mapping
idx_to_class = {
    0: "class_0",
    1: "class_1",
    2: "class_2",
    3: "class_3",
    4: "class_4"
}

# Define ARGS dictionary with all parameters
ARGS = {
    "concept_bank": concept_bank,
    "backbone_name": "clip:RN50",
    "idx_to_class": idx_to_class,
    "n_classes": 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PosthocLinearCBM — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PosthocLinearCBM output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 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 1 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

Concept Bottleneck Models (CBMs) map the inputs onto a set of interpretable concepts ("the bottleneck") and use the concepts to make predictions. A concept bottleneck enhances interpretability since it can be investigated to understand what concepts the model "sees" in an input and which of these concepts are deemed important. However, CBMs are restrictive in practice as they require dense concept annotations in the training data to learn the bottleneck. Moreover, CBMs often do not match the accuracy of an unrestricted neural network, reducing the incentive to deploy them in practice. In this work, we address these limitations of CBMs by introducing Post-hoc Concept Bottleneck models (PCBMs). We show that we can turn any neural network into a PCBM without sacrificing model performance while still retaining the interpretability benefits. When concept annotations are not available on the training data, we show that PCBM can transfer concepts from other datasets or from natural language descriptions of concepts via multimodal models. A key benefit of PCBM is that it enables users to quickly debug and update the model to reduce spurious correlations and improve generalization to new distributions. PCBM allows for global model edits, which can be more efficient than previous works on local interventions that fix a specific prediction. Through a model-editing user study, we show that editing PCBMs via conceptlevel feedback can provide significant performance gains without using data from the target domain or model retraining. The code for our paper can be found in https://github.com/mertyg/post-hoc-cbm.

For agents

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

get_harvested_code_for_paper("2205.15480")
get_code_for_paper("2205.15480")
have("2205.15480")

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

curl https://syntology.ai/api/ran/2205.15480.json

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

[![Syntology run record](https://syntology.ai/api/ran/2205.15480.svg)](https://syntology.ai/paper/2205.15480)

Connect an agent — have() is free.