- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 1
- Soundness
- 2
Read review
Paper summary
The paper proposes Tstar, a method for constructing training data for time-series models. It assigns windows to 18 predefined temporal archetypes using statistical fingerprints and rule-based labels, then uses this structural atlas to guide window selection and sampling, with optional targeted synthesis. Under matched training budgets, Tstar achieves lower forecasting errors than uniform-selection or uniform-sampling baselines for TinyTimeMixer and MOMENT-341M. A supporting study evaluates forecasting and imputation across six conventional architectures.
Critical strengths
Tstar improves forecasting performance over the uniform-selection and uniform-sampling baselines under matched training budgets. This suggests that allocating training exposure according to temporal structure can improve performance without increasing the number of training updates or the batch size.
Critical weaknesses
- The presentation makes the method harder to follow than necessary. The abstract and introduction use boldface so frequently that it becomes visually distracting and dilutes the emphasis on the main points. The contribution statements could also be more concise and easier to scan.
- The narrative references to Mix, Don't Pick read awkwardly. Referring to the authors, for example, "Nagpal et al. (2026)," could improve the flow of these sentences.
- Readers must also reconcile several overlapping concepts across the main text and appendices. These issues make it difficult to reconstruct the intervention and assess which components explain the reported gains.
Questions for authors
- The abstract states, "The atlas guides real-window selection and sampling." Could "atlas" be defined more concretely at first use?
- In Section 3.2, what does "a capped random sample of windows" mean, and what cap is applied?
- How are the 18 named temporal patterns defined?
2 comments
Response to Reviewer gFQs (Part 2 of 2: [Q2, Q3])
[Q2] Meaning of "A Capped Random Sample of Windows" and the Exact Cap
The cap applies only to our fast offline profiling step before training, limiting profiling to at most 16 series per file and 64 non-overlapping windows per series so huge files do not dominate the atlas counts; training itself uses the full data pool.
To prevent any confusion between offline profiling and actual pretraining, we will clearly separate them:
- Offline Pile profile: 96,020 profiled windows sampled under the 16 series/file and 64 windows/series cap from internal Pile files only (external TIME and FEV benchmarks never enter profiling).
- MOMENT-341M pretraining: samples from the full pool of 80,437,730 training crops across 2,087 Pile datasets.
- TinyTimeMixer pretraining: trains on all 197,787 matched real windows from 18 public sources.
[Q3] Plain-English Definitions of All 18 Named Temporal Patterns (TAT-18)
Each window is tested top-to-bottom from Priority 1 to 18 and assigned to the first matching pattern, so rare structural patterns are captured before common stationary noise.
| Priority & Pattern | Plain-English Definition |
|---|---|
| 1. Sparse binary | Mostly binary (0 or 1) or two-level discrete values |
| 2. Bursty count | Nonnegative integer counts with sharp bursts or spikes |
| 3. Count event | Nonnegative integer counts without sharp bursts |
| 4. Positive multiplicative | Strictly positive, non-unit-bounded series with strong multiplicative volatility and periodicity |
| 5. Bounded oscillator | Oscillatory signal confined within the unit interval [0, 1] |
| 6. Heavy-tail noise | Extreme tail spikes or high outlier rate |
| 7. Intermittent pulse | Mostly near-zero baseline with isolated pulses |
| 8. Heteroskedastic | Strong volatility change across halves without a mean level jump |
| 9. Smooth trend | Strong, steady upward or downward drift |
| 10. Piecewise trend | Upward or downward drift combined with a clear mean level shift |
| 11. Regime switch | Clear mean level jump with high autocorrelation |
| 12. Drifting period | Periodic signal whose dominant frequency drifts over time |
| 13. Strong single season | One sharp, dominant periodic cycle |
| 14. Multiple seasons | Multiple distinct periodic cycles |
| 15. Long-memory-like | Slowly decaying autocorrelation with persistent long-range dependence |
| 16. Random walk | High lag-1 persistence without a strong global linear trend |
| 17. Chaotic-like | Two fixed heuristic rules for irregular, low-periodicity dynamics (uncertified for physical chaos) |
| 18. Stationary stochastic | Fallback category assigned when rules 1 to 17 do not trigger |
For scaling, spectral, correlation, volatility, and shape descriptors are normalized or scale-invariant, while value-support checks (binary, integer, nonnegative, unit-interval [0, 1]) and the trend-slope guard inspect raw values so bounds, quantization, and exact zeros are preserved.
Thank you again for helping us make the paper much clearer and easier to follow. We specify the full method, TAT-18 rules, and exact configurations transparently and will release all code and manifests openly upon publication; if these clarifications and our new BLAST baseline results address your concerns, we would be sincerely grateful if you would raise score.
Response to Reviewer gFQs (Part 1 of 2: [W1, W2, W3, Q1])
Response to Reviewer gFQs (Part 1 of 2: [W1, W2, W3, Q1])
Thank you for showing us how to make the paper much clearer and easier to read. We will carefully fix every writing, formatting, and citation point you raised. During rebuttal, we also added a new 3-seed scratch comparison against BLAST (a strong feature-based balancing baseline), where our primary policy (Tstar-Broad) beats BLAST on all four TIME and FEV metrics (+5.29% to +8.97% paired mean task gain, winning 3/3 paired MASE seeds). Below we answer each point in plain English (Part 2 of 2 answers [Q2] and [Q3]).
[W1] Heavy Boldface and Concise Contributions
We will remove all inline boldface throughout the abstract and introduction (keeping headings only) and shorten our contributions into two scannable points:
[W2] Narrative Citations to "Mix, Don't Pick"
We will replace every awkward narrative title reference with standard author-year citations, "Nagpal et al. (2026)."
[W3] What Each Variant Means and Which Component Drives the Gains
The core idea is simple: raw time-series corpora are flooded with repetitive flat or noisy windows, so rebalancing real training windows with the 18-category atlas drives our main gains, while finer subclustering and synthesis are optional and exploratory.
To make the variants easy to separate on a first read, we will add a concise plain-English summary table showing what each variant does and what it achieves under fixed compute budgets:
In short, the simple real-window policy (Tstar-Broad on MOMENT and real-window selection on TTM) does the heavy lifting at zero extra pretraining updates, model parameters, or inference cost.
[Q1] Concrete Definition of "Atlas" at First Use
At first use, we will define "atlas" as a finite 18-category frequency table over fixed-length training windows that counts how often each temporal pattern occurs, so underrepresented patterns receive higher training exposure.
Standard sampling draws by raw volume and overexposes common stationary noise; the atlas shows which local dynamics are overrepresented and which rare dynamics need upweighting.