Underspecification does not imply Incoherence: The Risks of Semantic Collapse in Coding Models

• URL: http://arxiv.org/abs/2607.01953v1
• Date: Thu, 02 Jul 2026 09:43:15 GMT
• Title: Underspecification does not imply Incoherence: The Risks of Semantic Collapse in Coding Models
• Abstract summary: Large Language Models (LLMs) have become increasingly effective at generating code when task descriptions are clear and precise.<n>Yet, user-provided task descriptions are often ambiguous, incomplete, or contradictory, leaving critical aspects of the intended program behavior underspecified.<n>We find that LLMs frequently collapse onto a single incorrect interpretation of the task description, consistently generating coherent but behaviorally misaligned code.
• Score: 6.72873429875663
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Large Language Models (LLMs) have become increasingly effective at generating code when task descriptions are clear and precise. Yet, in practice, user-provided task descriptions are often ambiguous, incomplete, or contradictory, leaving critical aspects of the intended program behavior underspecified. In such cases, multiple behaviorally distinct interpretations may satisfy the description equally well, yet semantically differ in ways that matter/affect the user intent. A natural expectation, often assumed by researchers, is that prompt underspecification manifests as incoherence: When asked multiple times, an LLM produces multiple semantically distinct implementations reflecting the ambiguity of the task description. In this paper, we challenge this assumption. We find that LLMs frequently collapse onto a single incorrect interpretation of the task description, consistently generating coherent but behaviorally misaligned code. We term this failure mode detrimental semantic collapse and find that it affects over 10% of tasks in MBPP, 3% in HumanEval, and 32% of LiveCodeBench, all benchmarks assumed to be well-specified. By deliberately injecting underspecification issues in the benchmark prompts, the rate rises to over 5 times, exposing a fundamental blind spot in disambiguation and correctness estimation techniques that rely on incoherence as a proxy for prompt underspecification.

Related papers

• ClarifyCodeBench: Evaluating LLMs on Clarifying Ambiguous Requirements for Code Generation [54.788849448970154]
  We introduce ClarifyCodeBench, a novel interactive benchmark for evaluating Large Language Models' capability in resolving requirement ambiguity.<n>We conduct a systematic evaluation of six state-of-the-art LLMs using ClarifyCodeBench.<n>Our empirical results yield three critical insights: 1) Capability Decoupling: Strong code generation performance does not inherently translate to effective requirement clarification; 2) The Reasoning Paradox: While increased computational thinking enhances code correctness, it yields marginal gains in identifying ambiguities.
  arXiv  Detail & Related papers  (2026-07-01T09:58:59Z)
• Small Language Model Helps Resolve Semantic Ambiguity of LLM Prompt [13.863963355859175]
  We propose a pre-inference prompt optimization mechanism via explicit prompt disambiguation.<n>We identify semantic risks in the prompt, check their multi-perspective consistency, and resolve any semantic conflicts.<n>Our method improves reasoning performance by 2.5 points at a cost of only $0.02.
  arXiv  Detail & Related papers  (2026-04-25T12:13:26Z)
• Assessing the Impact of Requirement Ambiguity on LLM-based Function-Level Code Generation [14.755517753769837]
  We introduce Orchid, the first code generation benchmark specifically designed with ambiguous requirements.<n>It comprises 1,304 function-level tasks covering four distinct types of ambiguity: lexical, syntactic, semantic, and vagueness.<n>Our results demonstrate that ambiguity consistently degrades the performance of all evaluated LLMs, with the most pronounced negative effects observed in highly advanced models.
  arXiv  Detail & Related papers  (2026-04-23T10:07:33Z)
• Beyond Tokens: Semantic-Aware Speculative Decoding for Efficient Inference by Probing Internal States [14.780400014944926]
  Large Language Models (LLMs) achieve strong performance across many tasks but suffer from high inference latency due to autoregressive decoding.<n>We propose SemanticSpec, a semantic-aware speculative decoding framework that verifies entire semantic sequences instead of tokens.
  arXiv  Detail & Related papers  (2026-02-03T16:30:30Z)
• Hidden in Plain Sight: Reasoning in Underspecified and Misspecified Scenarios for Multimodal LLMs [28.913007638707427]
  Multimodal large language models (MLLMs) are increasingly deployed in open-ended, real-world environments.<n>This paper presents a systematic analysis of how current MLLMs handle implicit reasoning scenarios.<n>We find that models frequently fail to surface hidden issues, even when they possess the necessary perceptual and reasoning skills.
  arXiv  Detail & Related papers  (2025-05-30T21:47:28Z)
• A Framework for Evaluating LLMs Under Task Indeterminacy [49.298107503257036]
  Large language model (LLM) evaluations often assume there is a single correct response -- a gold label -- for each item in the evaluation corpus. We develop a framework for evaluating LLMs under task indeterminacy.
  arXiv  Detail & Related papers  (2024-11-21T00:15:44Z)
• Contrastive Instruction Tuning [61.97704869248903]
  We propose Contrastive Instruction Tuning to maximize the similarity between semantically equivalent instruction-instance pairs. Experiments on the PromptBench benchmark show that CoIN consistently improves LLMs' robustness to unseen instructions with variations across character, word, sentence, and semantic levels by an average of +2.5% in accuracy.
  arXiv  Detail & Related papers  (2024-02-17T00:09:32Z)
• Clarify When Necessary: Resolving Ambiguity Through Interaction with LMs [58.620269228776294]
  We propose a task-agnostic framework for resolving ambiguity by asking users clarifying questions. We evaluate systems across three NLP applications: question answering, machine translation and natural language inference. We find that intent-sim is robust, demonstrating improvements across a wide range of NLP tasks and LMs.
  arXiv  Detail & Related papers  (2023-11-16T00:18:50Z)
• Simple Linguistic Inferences of Large Language Models (LLMs): Blind Spots and Blinds [59.71218039095155]
  We evaluate language understanding capacities on simple inference tasks that most humans find trivial. We target (i) grammatically-specified entailments, (ii) premises with evidential adverbs of uncertainty, and (iii) monotonicity entailments. The models exhibit moderate to low performance on these evaluation sets.
  arXiv  Detail & Related papers  (2023-05-24T06:41:09Z)
• PROMPT WAYWARDNESS: The Curious Case of Discretized Interpretation of Continuous Prompts [99.03864962014431]
  Fine-tuning continuous prompts for target tasks has emerged as a compact alternative to full model fine-tuning. In practice, we observe a "wayward" behavior between the task solved by continuous prompts and their nearest neighbor.
  arXiv  Detail & Related papers  (2021-12-15T18:55:05Z)

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.