paper-with-me

Papers

Reasoning in Transformers - Mitigating Spurious Correlations and Reasoning Shortcuts

2024-03-17 · Daniel Enström, Viktor Kjellberg, Moa Johansson

Transformer language models are neural networks used for a wide variety of tasks concerning natural language, including some that also require logical reasoning. However, a transformer model may easily learn spurious patterns in the data, short-circuiting actual reasoning. In this paper we investigate to what extent transformers can be trained to a) approximate reasoning in propositional logic while b) avoiding known reasoning shortcuts via spurious correlations in the training data. To do so, we use a dataset with known spurious correlation between truth and e.g. the number of rules in the problem. We augment the data with proofs, and train two models: a generative transformer, WP-BART, trained on problems and their whole proofs, and a neuro-symbolic model, SIP-BART, trained on individual proof steps and combining the generative transformer model BART with a symbolic proof checker. We find that SIP-BART succeeds in avoiding reasoning shortcuts, while WP-BART does not. For SIP-BART, we then identify a few remaining reasoning errors, not previously described in the literature, arising from using a pre-trained language model. These are qualitatively analysed to create a taxonomy of four different types of additional pitfalls.

📄 PDF Abstract BibTeX arXiv:2403.11314

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage ModellingLogical Reasoning

Methods 이 논문이 사용한 방법론

Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Multi-Head Attention 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…

Similar Papers 제목 키워드 기반

The Trap of Trajectory: Towards Understanding and Mitigating Spurious Correlations in Agentic Memory

2026-05-10 · Luoxi Tang, Rupali Rajendra Vaje, Yuqiao Meng, Sakshi Sunil Narkar 외 arxiv

Agentic memory enables LLMs to persist information beyond a single context window and reuse it in later decisions, but it also introduces a new vulnerability: spurious correlations, where retrieved memory carries miscorr…

Mitigating Spurious Correlations Between Question and Answer via Chain-of-Thought Correctness Perception Distillation

2025-09-06 · Hongyan Xie, Yitong Yao, Yikun Ban, Zixuan Huang 외 arxiv

Large language models (LLMs) excel at reasoning tasks but are expensive to deploy. Thus small language models (SLMs) are fine-tuned on CoT data generated by LLMs to copy LLMs' abilities. However, these CoT data may inclu…

Causal Deep Q Network

2025-10-27 · Elouanes Khelifi, Amir Saki, Usef Faghihi arxiv

Deep Q Networks (DQN) have shown remarkable success in various reinforcement learning tasks. However, their reliance on associative learning often leads to the acquisition of spurious correlations, hindering their proble…

Reinforcement LearningCausal Inference

Mitigating Spurious Correlations in NLI via LLM-Synthesized Counterfactuals and Dynamic Balanced Sampling

2025-12-20 · Christopher Román Jaimes arxiv

Natural Language Inference (NLI) models frequently rely on spurious correlations rather than semantic reasoning. Existing mitigation strategies often incur high annotation costs or trigger catastrophic forgetting during …

Natural Language Inference

Correlation or Causation: Analyzing the Causal Structures of LLM and LRM Reasoning Process

2025-09-22 · Zhizhang FU, Guangsheng Bao, Hongbo Zhang, Chenkai Hu 외 arxiv

LLMs suffer from critical reasoning issues such as unfaithfulness, bias, and inconsistency, since they lack robust causal underpinnings and may rely on superficial correlations rather than genuine understanding. Successi…

Reinforcement Learning