When Choosing Plausible Alternatives, Clever Hans can be Clever

Kavumba, Pride; Inoue, Nozomu; Heinzerling, Benjamin; Singh, Keshav K.; Rutgeerts, Paul; Inui, Kentaro

doi:10.48550/arxiv.1911.00225

Cited by 3 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Manuscript received --/revised --1. https://en.wikipedia.org/wiki/Clever Hans [17] in their input? Are source code models "stochastic parrots" [18], that overly rely on memorization [19]?…”

Section: Introductionmentioning

confidence: 99%

What do pre-trained code models know about code?

Karmakar

Robbes

2021

2021 36th IEEE/ACM International Conference on Automated Software Engineering (ASE)

View full text Add to dashboard Cite

Pre-trained models of source code have recently been successfully applied to a wide variety of Software Engineering tasks; they have also seen some practical adoption in practice, e.g. for code completion. Yet, we still know very little about what these pre-trained models learn about source code. In this article, we use probing-simple diagnostic tasks that do not further train the models-to discover to what extent pre-trained models learn about specific aspects of source code. We use an extensible framework to define 15 probing tasks that exercise surface, syntactic, structural and semantic characteristics of source code. We probe 8 pre-trained source code models, as well as a natural language model (BERT) as our baseline. We find that models that incorporate some structural information (such as GraphCodeBERT) have a better representation of source code characteristics. Surprisingly, we find that for some probing tasks, BERT is competitive with the source code models, indicating that there are ample opportunities to improve source-code specific pre-training on the respective code characteristics. We encourage other researchers to evaluate their models with our probing task suite, so that they may peer into the hidden layers of the models and identify what intrinsic code characteristics are encoded.

show abstract

“…Manuscript received --/revised --1. https://en.wikipedia.org/wiki/Clever Hans [17] in their input? Are source code models "stochastic parrots" [18], that overly rely on memorization [19]?…”

Section: Introductionmentioning

confidence: 99%

What do pre-trained code models know about code?

Karmakar

Robbes

2021

2021 36th IEEE/ACM International Conference on Automated Software Engineering (ASE)

View full text Add to dashboard Cite

show abstract

“…masked token prediction) (Sun et al, 2020). However, interpreting and extracting that knowledge is difficult (requiring specialized probing tasks) and these models can be susceptible to exploiting superficial (Kavumba et al, 2019). CQA tasks could provide a unique opportunity to develop both explainable models (through producing causal explanation chains) and expand the reasoning capabilities of those models in QA settings.…”

Section: Task Examplementioning

confidence: 99%

“…Provided a premise, the goal is to select the most likely cause or effect from a pair of options. (Kavumba et al, 2019) introduced 500 additional training examples in Balanced-COPA to mitigate the corpus-level artifacts that were likely to be exploited by language models during finetuning.…”

Section: Calm-bench Tasksmentioning

confidence: 99%

CALM-Bench: A Multi-task Benchmark for Evaluating Causality-Aware Language Models

Dalal,

Buitelaar,

Arcan

2023

Findings of the Association for Computational Linguistics: EACL 2023

View full text Add to dashboard Cite

Causal reasoning is a critical component of human cognition and is required across a range of question-answering (QA) tasks (such as abductive reasoning, commonsense QA, and procedural reasoning). Research on causal QA has been underdefined, task-specific, and limited in complexity. Recent advances in foundation language models (such as BERT, ERNIE, and T5) have shown the efficacy of pre-trained models across diverse QA tasks. However, there is limited research exploring the causal reasoning capabilities of those language models and no standard evaluation benchmark. To unify causal QA research, we propose CALM-Bench, a multi-task benchmark for evaluating causality-aware language models (CALM). We present a standardized definition of causal QA tasks and show empirically that causal reasoning can be generalized and transferred across different QA tasks. Additionally, we share a strong multi-task baseline model which outperforms single-task fine-tuned models on the CALM-Bench tasks.

show abstract