Teaching Large Language Models When Not to Know: Learning Temporal Critique for Ex-Ante Reasoning

• URL: http://arxiv.org/abs/2605.14636v1
• Date: Thu, 14 May 2026 09:49:35 GMT
• Title: Teaching Large Language Models When Not to Know: Learning Temporal Critique for Ex-Ante Reasoning
• Abstract summary: Large language models (LLMs) often fail to reason under temporal cutoffs.<n>We propose TCFT, a framework that trains models to acquire cutoff-aware temporal verification.
• Score: 21.280999165642932
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models (LLMs) often fail to reason under temporal cutoffs: when prompted to answer from the standpoint of an earlier time, they exploit knowledge that became available only later. We study this failure through the lens of ex-ante reasoning, where a model must rely exclusively on information knowable before a cutoff. Through a systematic analysis of prompt-level interventions, we find that temporal leakage is highly sensitive to cutoff formulation and instruction placement: explicit cutoff statements outperform implicit historical framings, and prefix constraints reduce leakage more effectively than suffix constraints. These findings indicate that prompting can steer models into a temporal frame, but does not endow them with the ability to verify whether a response is temporally admissible. We further argue that supervised fine-tuning is insufficient, since ex-ante correctness is not an intrinsic property of an answer, but a relation between the answer and the cutoff. To address this gap, we propose TCFT, a Temporal Critique Fine-Tuning framework that trains models to acquire cutoff-aware temporal verification. Given a query, a cutoff, and a candidate response, TCFT teaches the model to identify post-cutoff leakage, explain temporal boundary violations, and judge temporal admissibility. Experiments with Qwen2.5-7B-Instruct and Qwen2.5-14B-Instruct show that TCFT consistently outperforms prompting and SFT baselines, reducing average leakage by 41.89 and 37.79 percentage points, respectively.

Related papers

• Understanding Why Language Models Hallucinate: Testing Reasoning Against Priors [11.702863087019809]
  Large language models often produce hallucinated answers that violate prompt-level constraints.<n>We study inference misalignment: a mismatch between the answer supported by the prompt and the answer favored by statistically salient latent associations.<n>We formalize this view with a latent key-task model, in which pretraining-frequency imbalance can cause a shortcut path to dominate the constraint-sensitive path.
  arXiv  Detail & Related papers  (2026-07-01T05:02:43Z)
• Temporal-Aware Reasoning Optimization for Video Temporal Grounding [55.29748680163419]
  We propose TaRO (Temporal-Aware Reasoning Optimization), a framework that explicitly enhances the model's ability of thinking with time.<n>First, we introduce a Constructive Reasoning Exploration that leverages pre-generated dense captions to construct reasoning paths grounded in explicit visual cues and timestamps.<n>Second, to evaluate reasoning quality, we design a Temporal-Sensitivity Reward. High-quality reasoning should be anchored to specific events and timestamps.
  arXiv  Detail & Related papers  (2026-06-08T09:21:01Z)
• A Progressive Training Strategy for Vision-Language Models to Counteract Spatio-Temporal Hallucinations in Embodied Reasoning [49.61652671596548]
  "Multi-image hallucination reasoning", where a massive performance drop between forward and temporal queries reveals a dependence on superficial shortcuts instead of genuine understanding.<n>To mitigate this, we develop a new dataset: Chain-of-temporal building on detailed reasoning into Chain steps and definitive judgments.<n>Our experiments demonstrate that this approach not only improves accuracy but also the forward-backward performance gap from over 70% to only 6.53%.
  arXiv  Detail & Related papers  (2026-04-12T07:48:44Z)
• Decoding Answers Before Chain-of-Thought: Evidence from Pre-CoT Probes and Activation Steering [5.427346259545067]
  Chain-of-thought (CoT) has become central to scaling reasoning capabilities in large language models.<n>We show that instruction-tuned models often determine their answer before generating CoT.
  arXiv  Detail & Related papers  (2026-03-02T04:33:55Z)
• All Leaks Count, Some Count More: Interpretable Temporal Contamination Detection in LLM Backtesting [5.719582970287213]
  We introduce a claim-level framework for detecting and quantifying this emphtemporal knowledge leakage<n>Our approach decomposes model rationales into atomic claims and categorizes them by temporal verifiability, then applies textitShapley values to measure each claim's contribution to the prediction.<n>TimeSPEC reduces Shapley-DCLR while preserving task performance, demonstrating that explicit, interpretable claim-level verification outperforms prompt-based temporal constraints for reliable backtesting.
  arXiv  Detail & Related papers  (2026-02-19T10:28:00Z)
• Negative-Aware Diffusion Process for Temporal Knowledge Graph Extrapolation [16.301114199423044]
  Temporal Knowledge Graph (TKG) reasoning seeks to predict future missing facts from historical evidence.<n>Negative-Aware Diffusion model for TKG Extrapolation (NADEx)<n>NADEx encodes subject-centric histories of entities, relations and temporal intervals into sequential embeddings.
  arXiv  Detail & Related papers  (2026-02-09T15:50:56Z)
• When Silence Is Golden: Can LLMs Learn to Abstain in Temporal QA and Beyond? [34.471723373734605]
  Large language models (LLMs) rarely admit uncertainty, often producing fluent but misleading answers, rather than abstaining (i.e., refusing to answer)<n>This weakness is even evident in temporal question answering, where models frequently ignore time-sensitive evidence and facts across different time-periods.<n>Existing approaches such as calibration might be unreliable in capturing uncertainty in complex reasoning.<n>We frame abstention as a teachable skill and introduce a pipeline that couples Chain-of-Thought (CoT) supervision with Reinforcement Learning (RL) guided by abstention-aware rewards.
  arXiv  Detail & Related papers  (2026-02-04T16:54:47Z)
• Simulated Ignorance Fails: A Systematic Study of LLM Behaviors on Forecasting Problems Before Model Knowledge Cutoff [8.706104833783318]
  Simulated Ignorance (SI) prompting models to suppress pre-cutoff knowledge has emerged as a potential solution.<n>Across 477 competition-level questions and 9 models, we find that SI fails systematically.<n>We conclude that RF on pre-cutoff events is methodologically flawed.
  arXiv  Detail & Related papers  (2026-01-20T08:21:55Z)
• Temporal Predictors of Outcome in Reasoning Language Models [0.0]
  Chain-of-thought (CoT) paradigm uses the elicitation of step-by-step rationales as a proxy for reasoning.<n>We show that, for harder questions, a drop in predictive accuracy highlights a selection artifact.<n>Overall, our results imply that for reasoning models, internal self-assessment of success tends to emerge after only a few tokens.
  arXiv  Detail & Related papers  (2025-11-03T08:57:18Z)
• LoGU: Long-form Generation with Uncertainty Expressions [49.76417603761989]
  We introduce the task of Long-form Generation with Uncertainty(LoGU)<n>We identify two key challenges: Uncertainty Suppression and Uncertainty Misalignment.<n>Our framework adopts a divide-and-conquer strategy, refining uncertainty based on atomic claims.<n>Experiments on three long-form instruction following datasets show that our method significantly improves accuracy, reduces hallucinations, and maintains the comprehensiveness of responses.
  arXiv  Detail & Related papers  (2024-10-18T09:15:35Z)
• Spurious Feature Eraser: Stabilizing Test-Time Adaptation for Vision-Language Foundation Model [86.9619638550683]
  Vision-language foundation models have exhibited remarkable success across a multitude of downstream tasks due to their scalability on extensive image-text paired data.<n>However, these models display significant limitations when applied to downstream tasks, such as fine-grained image classification, as a result of decision shortcuts''
  arXiv  Detail & Related papers  (2024-03-01T09:01:53Z)
• Mitigating Temporal Misalignment by Discarding Outdated Facts [58.620269228776294]
  Large language models are often used under temporal misalignment, tasked with answering questions about the present. We propose fact duration prediction: the task of predicting how long a given fact will remain true. Our data and code are released publicly at https://github.com/mikejqzhang/mitigating_misalignment.
  arXiv  Detail & Related papers  (2023-05-24T07:30:08Z)
• Generic Temporal Reasoning with Differential Analysis and Explanation [61.96034987217583]
  We introduce a novel task named TODAY that bridges the gap with temporal differential analysis. TODAY evaluates whether systems can correctly understand the effect of incremental changes. We show that TODAY's supervision style and explanation annotations can be used in joint learning.
  arXiv  Detail & Related papers  (2022-12-20T17:40:03Z)

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.