pith. sign in

arxiv: 2402.08939 · v3 · pith:L6DVLFH6new · submitted 2024-02-14 · 💻 cs.AI · cs.CL

Premise Order Matters in Reasoning with Large Language Models

classification 💻 cs.AI cs.CL
keywords reasoningorderingllmsorderpremiseperformanceaccuracybenchmark
0
0 comments X
read the original abstract

Large language models (LLMs) have accomplished remarkable reasoning performance in various domains. However, in the domain of reasoning tasks, we discover a frailty: LLMs are surprisingly brittle to the ordering of the premises, despite the fact that such ordering does not alter the underlying task. In particular, we observe that LLMs achieve the best performance when the premise order aligns with the context required in intermediate reasoning steps. For example, in deductive reasoning tasks, presenting the premises in the same order as the ground truth proof in the prompt (as opposed to random ordering) drastically increases the model's accuracy. We first examine the effect of premise ordering on deductive reasoning on a variety of LLMs, and our evaluation shows that permuting the premise order can cause a performance drop of over 30%. In addition, we release the benchmark R-GSM, based on GSM8K, to examine the ordering effect for mathematical problem-solving, and we again observe a significant drop in accuracy, relative to the original GSM8K benchmark.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Same Evidence, Different Answer: Auditing Order Sensitivity in Multimodal Large Language Models

    cs.CL 2026-06 unverdicted novelty 7.0

    All 18 audited MLLMs exhibit order sensitivity with per-facet flip rates of 24-50%, exceeding same-order decoder noise.

  2. ClassEval-Pro: A Cross-Domain Benchmark for Class-Level Code Generation

    cs.SE 2026-04 unverdicted novelty 7.0

    ClassEval-Pro benchmark shows frontier LLMs achieve at most 45.6% Pass@1 on class-level code tasks, with logic errors (56%) and dependency errors (38%) as dominant failure modes.

  3. PARTREP: Learning What to Repeat for Decoder-only LLMs

    cs.CL 2026-07 conditional novelty 6.0

    PartRep selects high-NLL tokens via a lightweight early-exit gate for partial prompt repetition, retaining most full-repetition gains at 59.4% KV cache and 79% prefill FLOPs on eight benchmarks.

  4. CaliDist: Calibrating Large Language Models via Behavioral Robustness to Distraction

    cs.LG 2026-06 unverdicted novelty 6.0

    CaliDist calibrates LLMs by scaling confidence according to how much predictions change under semantic distractors, cutting average ECE from 23% to 7% on seven NLU benchmarks across six models.

  5. Compared to What? Baselines and Metrics for Counterfactual Prompting

    cs.CL 2026-05 conditional novelty 6.0

    Counterfactual prompting effects on LLMs are often indistinguishable from those caused by meaning-preserving paraphrases, causing most previously reported demographic sensitivities to disappear under proper statistica...

  6. Hierarchical Reasoning Model

    cs.AI 2025-06 unverdicted novelty 5.0

    HRM is a recurrent architecture with high-level planning and low-level execution modules that reaches near-perfect accuracy on complex Sudoku, maze navigation, and ARC benchmarks using 27M parameters and 1000 samples ...