CoBe: A Benchmark For Conversational Counterfactual Text Editing

Arvind Raghavan; Inwoo Hwang; Shreyas Havaldar; Kai-Zhan Lee; Aurghya Maiti; Yushu Pan; Jeffrey Wu; Mingxuan Li

cs.CLOct 9, 2026

CoBe: A Benchmark For Conversational Counterfactual Text Editing

Authors: Arvind Raghavan, Inwoo Hwang, Shreyas Havaldar, Kai-Zhan Lee, Aurghya Maiti, Yushu Pan, Jeffrey Wu, Mingxuan Li, +1 more

Organizations: Causal Artificial Intelligence Lab Columbia University

Abstract

Counterfactual text editing is the ability to revise a given text in light of a hypothetical change while preserving facts not causally affected by that change. When asked what would have happened if a doctor had not recommended surgery, the model output should retain the patient's medical history and modify only the consequences of the decision. This requires causal understanding: a model must determine, without being explicitly instructed to, what events are affected by the hypothetical change and what should remain fixed. This skill is central to human reasoning, especially in safety-critical domains like healthcare or law. By contrast, the task of associational text generation merely generates text statistically associated with the altered condition. In the medical example, an associational response might make the patient's symptoms less severe because milder cases correlate with no surgery. Such responses produce a plausible scenario, but not the counterfactual version of the given one. Prior counterfactual-text datasets do not clearly separate these modes of reasoning. We introduce the Counterfactual Benchmark (CoBe), a testbed for models' zero-shot ability to counterfactually edit a given textual scenario when prompted to do so in a conversationally natural way. CoBe contains 2000 questions spanning healthcare, science and engineering, business, and everyday scenarios. We test a suite of frontier and smaller models, with frontier systems achieving an average score of $58.05\%$ . Our findings indicate that models frequently default to associational generation rather than preserving the causal structure of the original scenario. Moreover, performance does not scale consistently with either model size or output token length, suggesting that the failure may reflect model training objectives rather than insufficient model capacity.

Figures & tables

Figure 1: The three rungs of Pearl’s causal ladder are (i) associational; (ii) interventional; and (iii) counterfactual.

Figure 2: Counterfactual text editing, belonging to Rung 3 of Pearl’s causal ladder , keeps background factors constant , and reasons about what would have happened as the result of a hypothetical intervention . By contrast, associational text generation belongs to Rung 1, and generates alternative scenarios correlated with the altered condition (possibly changing background factors) . Note : counterfactual scenarios need not be realistic or likely to occur, ordinarily.

Figure 3: Idealized healthcare workflow depicting a staff reviewing medical records to generate case studies for training. Gemini 3.1 is prompted, in conversational language, to incorporate a counterfactual edit to patient treatment. Gemini incorrectly alters patient symptoms, causing downstream errors.

Prior work	Ctf-editing 3 steps	Narrative scenarios	No causal structure	Open-ended generation	Expert/formal verification
TimeTravel	✓	✓	✗	✓	✗
CRASS	✗	✓	✓	✗	✗
CLadder	✓	✗	✗	✗	✓
CounterBench	✓	✗	✗	✗	✓
CoBe (our work)	✓	✓	✓	✓	✓

Table 1: Comparison to prior benchmarks.

Figure 5

Figure 4: The types of underlying graphs that determine the interaction of variables in a story.

Figure 6: Failure rates across different evaluation criteria. E1-3 are described in Sec. 3.1 . E4 and E5 are interesting sub-types where LLMs reason wrongly about the direction of a downstream edit, and fail to correctly infer latent factors implicit in the story.

Figure 7: Accuracy across different graph types.

Figure 8: Accuracy across different query phrasings.

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Stages (1-4)	Output	Steps taken	Quality assurance and filtration
Counterfactual example collection	425 candidate counterfactual examples	A cohort of student researchers constructed examples following the dataset schema in Sec. 3. Each example contained an original scenario, a counterfactual edit request, and explicit evaluation criteria for LLM answers.	All contributors had completed a graduate causal inference course. Each example was required to specify a coherent counterfactual intervention, the resulting edit, and criteria identifying which facts should remain fixed and which consequences should change.
Core-set selection	106 diverse core examples crafted	The authors manually reviewed the 425 candidates and used them as a basis to craft a core set spanning diverse domains, event interactions, and causal structures.	We rejected candidates with ill-defined counterfactuals, ambiguous or unsound causal relationships, incoherent scenarios, or evaluation criteria that did not clearly distinguish valid from invalid edits.
Augmentation of core set	25 candidate variations per core example	Each core example was augmented using LLMs to produce variations across diverse themes and domains while preserving the original causal structure and logical relationships among events.	Generation was repeated until 25 satisfactory variations were obtained for each core example. We rejected and regenerated variations that were incoherent or contained unsound evaluation criteria.
Final verification and filtering	2,650 examples; 2,000 final examples retained after quality checks	The authors manually reviewed every set of 25 augmented examples and its associated evaluation criteria.	Every augmented set of 25 variations was checked for story coherence and for the correctness and clarity of its evaluation criteria. Only examples passing this review were included in the final dataset. External expert feedback led to additional examples being removed.

Appendix

Table 6: Dataset construction pipeline.

Figure 9: Accuracy across different failure modes.

Explore similar work

The Illusion of $\textit{What If}$ : Evaluating the Breakdown of Counterfactual Reasoning in LLMs
Aug 28, 2026Yucheng Wang, Yuetian Du, Zhengyi Liu +8Counterfactual Reasoning in Language Models Counterfactual Reasoning
Zhejiang University · Alibaba Group · City University of Hong Kong
Gumbel Machine: Counterfactual Student Writing Generation via Gumbel Noise Steering
May 26, 2026Hunter McNichols, Alexander Scarlatos, Mihai Dascalu +2Constrained Decoding Controllable Text Generation
University of Massachusetts Amherst · University Politehnica of Bucharest · Arizona State University
Compared to What? Baselines and Metrics for Counterfactual Prompting
May 1, 2026Zihao Yang, Mosh Levy, Yoav Goldberg +1Counterfactual Evaluation Causal Interventions in Language Models
Northeastern University · Bar-Ilan University · Allen Institute for AI

Aug 28, 2026J/K move · Enter open · S save

The Illusion of $\textit{What If}$ : Evaluating the Breakdown of Counterfactual Reasoning in LLMs

Yucheng Wang, Yuetian Du, Zhengyi Liu, Rongyu Zhang+7

Zhejiang University · Alibaba Group · City University of Hong Kong

Counterfactual Reasoning in Language Models Counterfactual Reasoning Counterfactual Evaluation Causal Reasoning in Language Models Faithfulness of Language Model Explanations

Open paper AskPDF arXiv

Counterfactual reasoning requires models to reason beyond the observed world and explain how altered conditions propagate through downstream consequences. Existing benchmarks largely target bounded settings with fixed variables or single gold outcomes, overlooking open-domain scenarios requiring causal-process evaluation. To this end, we present

\textbf{WhatIfBench}

, a diagnostic benchmark for open-domain, open-form, long-horizon counterfactual causal reasoning, containing 220 what-if questions across STEM, HSS, and Hybrid scenarios. To evaluate free-form responses, we further propose

\textbf{PRISM}

, which first converts each natural-language explanation into a Response-Derived Semantic Causal Graph of events, states, and mechanisms. On top of this graph, PRISM then jointly applies a Process Metric assessing graph-level causal validity and a Rubric Metric assessing answer-level explanatory adequacy. Evaluating six frontier LLMs with this framework, we find that WhatIfBench remains far from saturated: even the strongest model reaches only a 64.62% final score. Further analysis reveals persistent causal gaps, premise drift, and topology fragmentation, suggesting that fluent counterfactual narratives often mask fragile causal processes. The benchmark, code, and evaluation scripts are available at WhatIfBench.

Abstract

Explore similar work

The Illusion of What If\textit{What If}What If: Evaluating the Breakdown of Counterfactual Reasoning in LLMs

Gumbel Machine: Counterfactual Student Writing Generation via Gumbel Noise Steering

Compared to What? Baselines and Metrics for Counterfactual Prompting

The Illusion of $\textit{What If}$ : Evaluating the Breakdown of Counterfactual Reasoning in LLMs