Is the Future Predictable? Modeling Future Predictability for Reliable Temporal Reasoning

Yongyi Huang; Huaping Zhang; Wenyao Cui; Qiuchi Li; Lei Li; Baohua Zhang

cs.CLOct 8, 2026

Is the Future Predictable? Modeling Future Predictability for Reliable Temporal Reasoning

Authors: Yongyi Huang, Huaping Zhang, Wenyao Cui, Qiuchi Li, Lei Li, Baohua Zhang

Organizations: Beijing Institute of Technology · Zhongguancun Academy

Abstract

Large language models (LLMs) remain unreliable when reasoning about the future. Existing models often overlook the difference among future states, committing to specific answers even when the queried facts are inherently volatile. To capture this challenge, we introduce ChronoRegime, a future-oriented temporal reasoning task that requires models to determine whether historical temporal evidence justifies commitment to a future answer. Furthermore, we propose Confidence-Aware Temporal Reasoning (CTR), a framework that estimates future predictability from temporal dynamics and regulates answer commitment accordingly. Rather than performing unconditional forecasting, CTR commits only when the reconstructed evidence supports the target future state. Across multiple LLMs, CTR improves strict overall accuracy by 6.7 percentage points and reduces hallucination rates by 18.4 percentage points on average over the plain QA baseline. These results demonstrate that explicit predictability assessment is an effective mechanism for safer and more reliable temporal reasoning under future uncertainty.

Figures & tables

Figure 1: An example of how LLaMA3.1 performs in future-oriented temporal reasoning tasks. The model is unable to distinguish the predictable boundaries, and gives confident but incorrect answers at times that are inherently unpredictable.

Figure 2: Challenge of Current LLMs on future predictability modeling and the motivation of ChronoRegime in highlighting the needs for decision-making beyond unconditional prediction.

Figure 3: Illustration of the boundary-probing setup. The change point acts as a predictability boundary: before it, the anchor answer is supported by temporal evidence; after it, the model must revise its commitment or avoid unsupported prediction.

Model	PreAcc $\uparrow$	PostAcc $\uparrow$	CAS $\uparrow$	Post-PR $\downarrow$	WCCR $\downarrow$	$\Delta H$
LLaMA	0.500	0.158	0.000	0.429	0.832	0.006
Qwen3	0.911	0.219	0.250	0.924	0.786	0.002
Mistral	0.333	0.082	0.036	0.276	0.867	0.001
GPT-4o	0.839	0.316	0.286	0.689	0.643	0.022

Table 1: Boundary-probing results.

Figure 4: Average answer entropy around factual change points. Across models, uncertainty remains largely flat before and after temporal transitions, indicating that current LLMs do not reliably increase uncertainty when future commitment becomes less justified.

Figure 5: Overview for Confidence-Aware Temporal Reasoning (CTR). CTR leverages entropy-based confidence to shift from static retrieval to dynamic evolution modeling. Notably, its confidence-aware mechanism suppresses erroneous historical evidence (e.g., the low-confidence 2011 entry) to mitigate error propagation. By inducing explicit evolution patterns, CTR overcomes the persistence bias seen in the traditional reasoning modes and prioritizes reliability, avoiding hallucinated extrapolation for volatile future queries.

Model	Variant	Answer Accuracy (%) $\uparrow$				Type Acc. (%) $\uparrow$	Halluc. (%) $\downarrow$
Model	Variant	Fast	Slow	Never	Overall
LLaMA3.1	Base	6.9	21.0	50.0	26.7	23.7	36.8
	+Controlled Retrieval	5.9	29.4	53.4	30.6	26.1	33.2
	+CTR	6.9	31.9	56.0	32.6	55.3	22.6
Qwen3	Base	4.9	26.9	41.4	25.2	35.0	56.7
	+Controlled Retrieval	8.8	21.0	47.4	26.4	54.9	54.6

Table 2: Main results (%) on answer accuracy, type accuracy, and hallucination rate. Controlled Retrieval reasoning yields inconsistent gains over plain QA, while CTR consistently improves performance on slow- and never-changing facts and reduces hallucinations across models.

Method	Acc. $\uparrow$	Halluc. $\downarrow$
GPT-4o	44.8	17.8
Entropy-Gated	41.2	10.4
Heterogeneous Plurality	42.7	28.2
5-sample Self-Consistency	52.8	26.7
Controlled Retrieval	47.8	13.7
CTR	53.7	11.3

Table 3: Comparison with alternative uncertainty- and agreement-based baselines on GPT-4o. All methods use the same hallucination evaluation protocol.

	GPT-4o			Qwen3
Ablation	$\Delta$ Hall. $\downarrow$	$\Delta$ Cov.	$\Delta$ Acc. $\uparrow$	$\Delta$ Hall. $\downarrow$	$\Delta$ Cov.	$\Delta$ Acc. $\uparrow$
w/o Entropy	+7.4	-2.7	-0.6	+0.9	+22.8	-3.9
w/o Singleton Filtering	+7.7	-1.5	-3.9	-1.5	+21.1	-3.0
w/o Change-Point Modeling	+5.6	+0.3	-0.4	-1.5	+24.9	-4.5
Timeline-only	+10.1	+3.0	-5.5	+9.2	+14.2	-1.4
w/o Commitment Regulation	+5.9	+25.2	-11.6	+27.6	+40.9	-8.3

Table 4: Component ablations relative to full CTR. Positive

\Delta

Hall. indicates increased hallucination. All values are percentage-point changes.

Figure 6: Abstention performance of CTR policy.

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 7: Prediction flip curves around factual change points. The blue curve shows the probability of predicting the pre-change anchor answer, while the orange curve shows the probability of predicting the post-change answer. A boundary-aware model should reduce anchor-answer predictions and increase post-change predictions after the change point.

Case	Model	Gold	Controlled Retrieval	Confidence-Aware TR (CTR)
The number of books published by Shoshana Zuboff	GPT-4o	3	Answer: 4 Reasoning: Assumes a slow publication rate without citing concrete historical evidence.	Answer: 3 Reasoning: "The number of books remained stable at 3 from 2018 to 2024. Given the Slow-changing pattern, it is likely unchanged in 2025."
Children of Leonardo DiCaprio	Qwen3	0	Answer: 1 Reasoning: "Leonardo DiCaprio has been publicly known to have one child, a daughter named Matilda, born in 2004."( Hallucination )	Answer: 0 Reasoning: "The evidence indicates a stable period from 1978 to 2024 with no children."

Appendix

Table 5: Two representative case studies comparing the controlled retrieve-then-reason baseline with our CTR framework. The baseline exhibits speculative extrapolation (Case 1) and factual hallucination (Case 2), whereas CTR produces more reliable predictions through structured temporal grounding.

Figure 8: An illustrative case showing how CTR suppresses low-confidence change points and commits to future predictions only when long-term temporal stability is observed.

Fact Type	Query ( $q_{i}$ )	Ref. Time ( $t_{0}^{i}$ )	Current Answer ( $a_{0}^{i}$ )	Next Answer ( $a_{1}^{i}$ )
Never	X won the 2020 Formula 1 world driver’s championship.	2022	Lewis Hamilton	N/A
Never	The real name of the Unabomber was X.	2022	Theodore John Kaczynski	N/A
Slow	The Winter Olympics were held most recently in X.	2022	Beijing, China	Italy
Slow	The next leap year is X.	2024	2028	2032
Fast	The most recent movie in the Marvel Cinematic Universe is X.	2025	The Fantastic Four: First Steps	Spider-Man: Brand New Day
Fast	King Gizzard’s most recent studio album is X.	2025	Phantom Island	N/A

Appendix

Table 6: Representative dataset examples. The Next Answer field is used only for evaluation and is not provided to the model. N/A indicates that no annotated future transition answer is provided; its interpretation depends on the evolution type.

Temporal Reasoning in Language Models

Temporal reasoning in language models studies how models represent, order, update, and apply facts and events across time.

See all 99 papers

Forecasting With LLMs: Improved Generalization Through Feature Steering
Jun 25, 2026Humzah Merchant, Bradford LevyLook-Ahead Bias in Language Models LLM Interpretability
University of Chicago, Illinois, United States.
Forecasting Future Behavior as a Learning Task
Jun 9, 2026Mosh Levy, Yoav Goldberg, Asa Cooper SticklandLLM Interpretability Large Reasoning Models
Bar-Ilan University · Constellation · UK AI Security Institute +1
Monitoring the Internal Monologue: Probe Trajectories Reveal Reasoning Dynamics
May 18, 2026Maciej Chrabąszcz, Aleksander Szymczyk, Marcin Sendera +2Language Model Safety Evaluation LLM Interpretability
NASK - National Research Institute, Poland · Faculty of Electronics and Information Technology, Warsaw University of Technology, Poland · Faculty of Mathematics and Computer Science, Jagiellonian University, Poland +3

Jun 25, 2026J/K move · Enter open · S save

Forecasting With LLMs: Improved Generalization Through Feature Steering

Humzah Merchant, Bradford Levy

University of Chicago, Illinois, United States.

Look-Ahead Bias in Language Models LLM Interpretability Temporal Reasoning in Language Models Language Model Steering LLM-Based Time Series Forecasting Causal Interventions in Language Models+1

Open paper AskPDF arXiv

Successful forecasting involves identifying patterns between historical and future states of the world which generalize to future observations. We apply LLMs to a variety of forecasting tasks and inspect their internal states using sparse autoencoders to understand whether they appear to rely on time-specific pieces of knowledge versus generalizable patterns. Our analyses identify features associated with both time-aware reasoning and look-ahead-biased reasoning. We then apply the LLMs to an entirely different domain and intervene on these features. We find that amplifying time-awareness features substantially reduces look-ahead bias on forecasting prompts while preserving general reasoning performance. In contrast, steering the candidate look-ahead-bias features does not produce an effect. These results suggest that interpretable temporal features can be used to causally shift LLMs toward more historically grounded reasoning.

Abstract

Temporal Reasoning in Language Models

Explore similar work

Forecasting With LLMs: Improved Generalization Through Feature Steering

Forecasting Future Behavior as a Learning Task

Monitoring the Internal Monologue: Probe Trajectories Reveal Reasoning Dynamics