Event Signature Transfer: Model-Agnostic Forecast Scenario Construction from Historical Events

Karthik Sridhar; Aaditya Jain; Murari Mandal; Saurabh Deshpande

cs.AISep 19, 2026

Event Signature Transfer: Model-Agnostic Forecast Scenario Construction from Historical Events

Authors: Karthik Sridhar, Aaditya Jain, Murari Mandal, Saurabh Deshpande

Organizations: Birla AI Labs · KIIT Bhubaneswar

Abstract

Forecasters often know an event is imminent but not the shape, size, or timing of its effect. We introduce Event Signature Transfer (EST), a training-free, model-agnostic operator that turns a completed past event into an explicit forecast scenario. EST removes a source event's own trend and seasonality, then scales and retimes the remaining event signature onto a native forecast, preserving the forecast's linked structure and reducing to it exactly at zero strength. Because it reads only output quantiles, EST applies to any quantile forecaster, with no training, no model internals, at transfer time. Across twelve real episodes and ten synthetic scenarios on Chronos-2, TimesFM-2.5 and Toto-2.0, manually configured EST reduces real-episode WQL by 21.7-90% in-sample. On Chronos-2, it leads eleven of twelve matched comparisons against covariate conditioning, activation editing and raw replay. The operator builds a scenario; it does not estimate its likelihood.

Figures & tables

Panel of Figure 1 — Figure 1: Event Signature Transfer Framework. Step 1 : on a scale where relative changes are comparable, remove the source’s usual trend and seasonal pattern, leaving the event signature. Step 2 : a frozen TSFM produces the native target forecast $\mathbf{Q}^{0}$ . Step 3 : choose strength and timing, apply the signature to every quantile of $\mathbf{Q}^{0}$ , and map back to the target’s scale, giving the steered output.

Panel of Figure 2 — Figure 2: Source, target, and result. (a) An authored dip preserves the target’s native rhythm. (b) The May 2009 Yampa signature steers the May 2023 forecast toward the observed rise (85.6% in-sample WQL reduction). Gold denotes the source event and projected background.

Episode	Covariate	t2t-C2 †	Replay	EST
Election attention	+14.1	0.0	+56.4	+72.7
Gasoline / Harvey	-1.4	+72.9	+80.4	+88.6
Refinery / Harvey	+2.7	+10.3	-140.3	+70.2
Thanksgiving travel	+11.1	+27.9	+15.3	+54.4
Oil / OPEC	-13.4	+40.4	+24.1	+61.5
NYC smoke	+2.4	+2.1	+29.7	+78.7

Table 1: Chronos-2 in-sample WQL skill (%) relative to the untouched

\mathbf{Q}^{0}

forecast. Higher is better; zero matches

\mathbf{Q}^{0}

. Shading marks best and runner-up . Constructions and per-backbone results: Appendix C .

\dagger

time2time-inspired adaptation.

Panel of Figure 3 — Figure 3: The 2020 election response steers the 2024 election forecast. Forecast starts 22 October, before the scheduled 5 November election. Logarithmic axes; native $\mathbf{Q}^{0}$ dashed, steered medians solid. Independent in-sample scenarios reduce WQL by 72.7/64.1/67.9% for Chronos-2/TimesFM/Toto (Tables A6 – A8 ).

Appendix figures & tables17 assets

Supplementary material from the paper’s appendix.

Appendix

Link	$G(v)$	Target change	Within-step preservation
Log	$\log v$	Multiply by $\exp(e)$	Quantile ratios
Relative	$v/s$ , $s>0$	Add $se$	Raw differences
Logit	$\log\frac{v-a}{b-v}$	Multiply odds by $\exp(e)$	Log-odds differences

Appendix

Table A1: The link determines what a transferred change means. Source and target may use different links.

Episode and measurement	Analogue and rationale
Election attention Wikimedia pageviews	2020 election article for 2024. A known election date motivates a concentrated attention pulse.
Gasoline / Harvey EIA/FRED GASALLW	Katrina (2005) for Harvey (2017): disrupted supply raises retail gasoline prices [ 17 , 9 ] .
Refinery / Harvey EIA WPULEUS3	Ike (2008) for Harvey (2017): refinery outage and recovery, opposite in direction to gasoline [ 17 ] .
Thanksgiving travel TSA checkpoint counts	Thanksgiving 2023 for 2024: the published holiday calendar motivates a trough and rebound.
Oil / OPEC EIA/FRED DCOILWTICO	November 2014 no-cut meeting for December 2015: a sustained price-decline scenario; the policy outcome is uncertain.
NYC smoke EPA AirData, NYC	July 2021 transported smoke for June 2023: a pollution pulse, with weather and source-mixture confounds [ 10 ] .

Appendix

Table A2 : Real measurements and the response each analogue is intended to supply. Provider names link to data sources.

Episode	Source event	Target forecast	$N_{x}/N_{y}$	$E/H$
Election attention	2020-10-20 2020-11-20	2024-10-22 2024-11-22	28/512	32/32
Gasoline / Harvey	2005-08-29 2005-09-19	2017-08-28 2017-09-18	156/256	4/4
Refinery / Harvey	2008-09-12 2008-11-28	2017-08-25 2017-11-10	156/260	12/12
Thanksgiving travel	2023-11-10 2023-12-11	2024-11-15 2024-12-16	42/336	32/32
Oil / OPEC	2014-11-28 2015-01-02	2015-12-04 2016-01-08	225/225	24/24
NYC smoke	2021-07-19 2021-07-26	2023-06-05 2023-06-12	413/512	8/8

Appendix

Table A3 : Inclusive event and forecast dates.

N_{x}/N_{y}

: source/target history lengths;

E/H

: source-event/forecast lengths. Histories end immediately before their respective windows. Oil and tax steps are observed business/trading days; the other clocks are daily or weekly as supplied.

Episode	Source / target link	$P$	Trend	$k$	$\tau$	$s_{x}/s_{y}$
Election attention	log / log	7	linear	1	0.04	–
Gasoline / Harvey	log / log	52	linear	1	0.04	–
Refinery / Harvey	log / logit	52	linear	3	0.04	–
Thanksgiving travel	log / log	7	linear	1	0.04	–
Oil / OPEC	log / log	0	linear	3	0.04	–
NYC smoke	log / log	0	linear	1	0.04	–

Appendix

Table A4 : Default source extraction. TimesFM ILI instead uses the optional background in Appendix C.1 .

P

: seasonal period (zero omits it);

k

: smoothing width;

\tau

: shape tolerance;

s_{x}/s_{y}

: relative-link scales in original units. No explicit knots or caps are used.

Panel of Figure 9 — Figure A1 : Twelve source response shapes. Grey is the raw background-adjusted residual; gold is the extracted signature. The virtual zero anchor is at step $-1$ . Axes are in each episode's source-link units and have independent scales. These are source responses, not target truth or causal effects. ILI shows the default signature used by Chronos-2 and Toto; the optional TimesFM background is described in Appendix C.1 .

Scenario	Shape	$b_{y}$	$a_{y}/a_{x}$	$w/\ell/D$
Downward crash	crash	0	0/0	1/0/32
Upward rise	rise	-0.002	0/0	1.2/0/32
Downward transient shock and recovery	down spike	0	0/0	0.8/0/32
Upward transient surge and recovery	up spike	0	0/0	1.1/0/32
Crash with the target’s weekly rhythm retained	crash	0.0005	0.1/0	0.9/0/32
Surge with the target’s weekly rhythm retained	up spike	0	0.12/0	1/0/32

Appendix

Table A5 : Authored synthetic data and scenario controls.

b_{y}

is target slope,

a_{y}/a_{x}

the target/source seasonal amplitudes.

w/\ell/D

gives strength/delay/duration.

Episode	Covariate	t2t-C2 †	Replay	EST
Election attention	+14.1	0.0	+56.4	+72.7
Gasoline / Harvey	-1.4	+72.9	+80.4	+88.6
Refinery / Harvey	+2.7	+10.3	-140.3	+70.2
Thanksgiving travel	+11.1	+27.9	+15.3	+54.4
Oil / OPEC	-13.4	+40.4	+24.1	+61.5
NYC smoke	+2.4	+2.1	+29.7	+78.7

Appendix

Table A6 : Chronos-2: in-sample WQL skill (%). Best and runner-up are shaded within each row. The native forecast

\mathbf{Q}^{0}

has zero skill.

\dagger

time2time-inspired adaptation.

Episode	Replay	EST
Election attention	+52.5	+64.1
Gasoline / Harvey	+79.1	+84.8
Refinery / Harvey	-103.9	+67.9
Thanksgiving travel	+19.7	+46.5
Oil / OPEC	+42.4	+68.6
NYC smoke	+24.7	+77.5

Appendix

Table A7 : TimesFM 2.5: in-sample WQL skill (%). Best and runner-up are shaded within each row. The native forecast

\mathbf{Q}^{0}

has zero skill. Covariate and activation interfaces were not evaluated.

Episode	Replay	EST
Election attention	+52.3	+67.9
Gasoline / Harvey	+80.6	+85.6
Refinery / Harvey	-113.5	+68.9
Thanksgiving travel	+19.5	+49.1
Oil / OPEC	+45.0	+65.8
NYC smoke	+27.4	+74.5

Appendix

Table A8 : Toto 2.0: in-sample WQL skill (%). Best and runner-up are shaded within each row. The native forecast

\mathbf{Q}^{0}

has zero skill. Covariate and activation interfaces were not evaluated.

Panel of Figure 14 — Figure A2 : Complete real forecast horizons on all three backbones. Blue dashed: $\mathbf{Q}^{0}$ ; green solid: EST; black: observations. Shading: q10–q90. Each model uses its own reported scenario.

Panel of Figure 15 — Figure A3 : Complete real forecast horizons on all three backbones. Blue dashed: $\mathbf{Q}^{0}$ ; green solid: EST; black: observations. Shading: q10–q90. Each model uses its own reported scenario.

Panel of Figure 16 — Figure A4 : Complete real forecast horizons on all three backbones. Blue dashed: $\mathbf{Q}^{0}$ ; green solid: EST; black: observations. Shading: q10–q90. Each model uses its own reported scenario.

Panel of Figure 17 — Figure A5 : Synthetic scenarios: source, signature and full target history above; Chronos-2, TimesFM and Toto forecasts below. No target future truth or accuracy score is defined.

Panel of Figure 18 — Figure A6 : Synthetic scenarios: source, signature and full target history above; Chronos-2, TimesFM and Toto forecasts below. No target future truth or accuracy score is defined.

Panel of Figure 19 — Figure A7 : Synthetic scenarios: source, signature and full target history above; Chronos-2, TimesFM and Toto forecasts below. No target future truth or accuracy score is defined.

Panel of Figure 20 — Figure A8 : Synthetic scenarios: source, signature and full target history above; Chronos-2, TimesFM and Toto forecasts below. No target future truth or accuracy score is defined.

Panel of Figure 21 — Figure A9 : Synthetic scenarios: source, signature and full target history above; Chronos-2, TimesFM and Toto forecasts below. No target future truth or accuracy score is defined.

Explore similar work

WorldReasoner: Evaluating Whether Language Model Agents Forecast Events with Valid Reasoning
Jun 10, 2026Yizhou Chi, Eric Chamoun, Zifeng Ding +1Event Time Prediction LLM Agent Evaluation
Department of Computer Science and Technology University of Cambridge
Forecast-Dojo: Replayable Environments for Benchmarking and Training LLM Forecasting Agents
Sep 24, 2026Liqin Ye, Haorui Wang, Fardin Ahmed +8LLM Agent Evaluation Forecasting Benchmarks
Georgia Institute of Technology · Amazon · University of Florida
From Narrative to Auditable Forecasts: A Structured Scaffold for Agentic Forecasting
Sep 5, 2026Yuanpu Cao, Yongkang Du, Yurui Chang +2Probabilistic Forecasting
The Pennsylvania State University

Jun 10, 2026, cs.CLJ/K move · Enter open · S save

WorldReasoner: Evaluating Whether Language Model Agents Forecast Events with Valid Reasoning

Yizhou Chi, Eric Chamoun, Zifeng Ding, Andreas Vlachos

Department of Computer Science and Technology University of Cambridge

Event Time Prediction LLM Agent Evaluation Forecasting Benchmarks Probabilistic Forecasting Temporal Data Leakage

Open paper AskPDF arXiv

Forecasting real-world events requires language-model agents to reason under uncertainty from incomplete, time-bounded information. Yet evaluating whether agents genuinely forecast requires more than final-answer accuracy: a model may be correct by recalling memorized training facts, citing fabricated evidence, or producing an unsupported causal story. We present WorldReasoner, an evaluation framework for temporally valid event forecasting. Each task gives an agent a resolved forecasting question, a simulated forecast date, and access only to evidence available before that date; after resolution, the framework scores the submitted probability, cited evidence, and optional causal event graph. WorldReasoner reports three complementary axes: outcome quality against resolved answers, evidence quality over cited sources, and reasoning quality against post-resolution hindsight graphs. The benchmark is built by an agentic construction pipeline that generates forecasting questions, collects time-stamped evidence, and builds hindsight reference graphs at scale, yielding 345 resolved tasks derived from 14,141 articles with graphs covering 8,087 extracted events. Across six controlled agent settings, temporally valid retrieval is the strongest driver of outcome accuracy; causal graph construction improves key-event recovery; and correct graph-enabled forecasts are more strongly grounded in key events and relevant sources, yet agents still struggle to convert grounded evidence into calibrated probabilities.

Abstract

Explore similar work

WorldReasoner: Evaluating Whether Language Model Agents Forecast Events with Valid Reasoning

Forecast-Dojo: Replayable Environments for Benchmarking and Training LLM Forecasting Agents

From Narrative to Auditable Forecasts: A Structured Scaffold for Agentic Forecasting