Focus on Your Question! Interpreting and Mitigating Toxic CoT Problems in Commonsense Reasoning

• URL: http://arxiv.org/abs/2402.18344v2
• Date: Thu, 27 Jun 2024 06:54:58 GMT
• Title: Focus on Your Question! Interpreting and Mitigating Toxic CoT Problems in Commonsense Reasoning
• Authors: Jiachun Li, Pengfei Cao, Chenhao Wang, Zhuoran Jin, Yubo Chen, Daojian Zeng, Kang Liu, Jun Zhao,
• Abstract summary: Large language models exhibit high-level commonsense reasoning abilities. CoT-like methods lead to a considerable number of originally correct answers turning wrong. We use attribution tracing and causal tracing methods to probe the internal working mechanism of the model.
• Score: 21.951313919964484
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models exhibit high-level commonsense reasoning abilities, especially with enhancement methods like Chain-of-Thought (CoT). However, we find these CoT-like methods lead to a considerable number of originally correct answers turning wrong, which we define as the Toxic CoT problem. To interpret and mitigate this problem, we first utilize attribution tracing and causal tracing methods to probe the internal working mechanism of the LLM during CoT reasoning. Through comparisons, we prove that the model exhibits information loss from the question over the shallow attention layers when generating rationales or answers. Based on the probing findings, we design a novel method called RIDERS (Residual decodIng and sERial-position Swap), which compensates for the information deficit in the model from both decoding and serial-position perspectives. Through extensive experiments on multiple commonsense reasoning benchmarks, we validate that this method not only significantly eliminates Toxic CoT problems (decreased by 23.6%), but also effectively improves the model's overall commonsense reasoning performance (increased by 5.5%).

Related papers

• Towards Robust Text Classification: Mitigating Spurious Correlations with Causal Learning [2.7813683000222653]
  We propose the Causally Calibrated Robust ( CCR) to reduce models' reliance on spurious correlations. CCR integrates a causal feature selection method based on counterfactual reasoning, along with an inverse propensity weighting (IPW) loss function. We show that CCR state-of-the-art performance among methods without group labels, and in some cases, it can compete with the models that utilize group labels.
  arXiv  Detail & Related papers  (2024-11-01T21:29:07Z)
• Causal Order Discovery based on Monotonic SCMs [5.47587439763942]
  We introduce a novel sequential procedure that directly identifies the causal order by iteratively detecting the root variable. This method eliminates the need for sparsity assumptions and the associated optimization challenges. We demonstrate the effectiveness of our approach in sequentially finding the root variable, comparing it to methods that maximize Jacobian sparsity.
  arXiv  Detail & Related papers  (2024-10-24T03:15:11Z)
• Unveiling the Statistical Foundations of Chain-of-Thought Prompting Methods [59.779795063072655]
  Chain-of-Thought (CoT) prompting and its variants have gained popularity as effective methods for solving multi-step reasoning problems. We analyze CoT prompting from a statistical estimation perspective, providing a comprehensive characterization of its sample complexity.
  arXiv  Detail & Related papers  (2024-08-25T04:07:18Z)
• A Hopfieldian View-based Interpretation for Chain-of-Thought Reasoning [48.51969964676017]
  Chain-of-Thought (CoT) holds a significant place in augmenting the reasoning performance for large language models. We propose a Read-and-Control approach for controlling the accuracy of CoT.
  arXiv  Detail & Related papers  (2024-06-18T04:07:13Z)
• Towards Faithful Chain-of-Thought: Large Language Models are Bridging Reasoners [19.40385041079461]
  Large language models (LLMs) suffer from serious unfaithful chain-of-thought (CoT) issues. We first study the CoT faithfulness issue at the granularity of CoT steps, identify two reasoning paradigms. We then conduct a joint analysis of the causal relevance among the context, CoT, and answer during reasoning.
  arXiv  Detail & Related papers  (2024-05-29T09:17:46Z)
• ChainLM: Empowering Large Language Models with Improved Chain-of-Thought Prompting [124.69672273754144]
  Chain-of-Thought (CoT) prompting can enhance the reasoning capabilities of large language models (LLMs) Existing CoT approaches usually focus on simpler reasoning tasks and thus result in low-quality and inconsistent CoT prompts. We introduce CoTGenius, a novel framework designed for the automatic generation of superior CoT prompts.
  arXiv  Detail & Related papers  (2024-03-21T11:34:26Z)
• ER: Equivariance Regularizer for Knowledge Graph Completion [107.51609402963072]
  We propose a new regularizer, namely, Equivariance Regularizer (ER) ER can enhance the generalization ability of the model by employing the semantic equivariance between the head and tail entities. The experimental results indicate a clear and substantial improvement over the state-of-the-art relation prediction methods.
  arXiv  Detail & Related papers  (2022-06-24T08:18:05Z)
• Tesseract: Tensorised Actors for Multi-Agent Reinforcement Learning [92.05556163518999]
  MARL exacerbates matters by imposing various constraints on communication and observability. For value-based methods, it poses challenges in accurately representing the optimal value function. For policy gradient methods, it makes training the critic difficult and exacerbates the problem of the lagging critic. We show that from a learning theory perspective, both problems can be addressed by accurately representing the associated action-value function.
  arXiv  Detail & Related papers  (2021-05-31T23:08:05Z)
• Detached Error Feedback for Distributed SGD with Random Sparsification [98.98236187442258]
  Communication bottleneck has been a critical problem in large-scale deep learning. We propose a new distributed error feedback (DEF) algorithm, which shows better convergence than error feedback for non-efficient distributed problems. We also propose DEFA to accelerate the generalization of DEF, which shows better bounds than DEF.
  arXiv  Detail & Related papers  (2020-04-11T03:50:59Z)
• Towards Out-of-Distribution Detection with Divergence Guarantee in Deep Generative Models [22.697643259435115]
  Deep generative models may assign higher likelihood to out-of-distribution (OOD) data than in-distribution (ID) data. We prove theorems to investigate the divergences in flow-based model. We propose two group anomaly detection methods.
  arXiv  Detail & Related papers  (2020-02-09T09:54:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.