Pith. sign in

REVIEW 2 major objections 5 minor 44 references

Conformal Coverage Guarantees for Any Video Temporal Grounder

T0 review · 2 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A post-hoc conformal wrapper gives any video temporal grounder a finite-sample coverage guarantee.

desk verdict Solid split-conformal wrapper for temporal grounding, with one real hole: the two-parameter variant is not proven to cover at 1-alpha, so its efficiency comparisons are at unmatched coverage. read the letter →

arxiv 2608.07434 v1 pith:HRUITLIW submitted 2026-08-07 cs.CV cs.IR

Aseel Mohamed , Rasul Khanbayov , Erchin Serpedin , Hasan Kurban This is my paper · ORCID
classification cs.CVcs.IR
keywords videotemporalgroundingconformalpredictiondistribution-freecoverageuncertaintyquantificationblack-boxmodelnonconformityscoreregioncalibrationwrapper
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that any fixed video temporal grounder—whether a trained localizer or a black-box video–language model—can be wrapped post-hoc so that its output becomes a temporal region containing the true moment with probability at least $1-\alpha$. The wrapper calibrates a single scalar, the quantile of a temporal nonconformity score over held-out labels, then widens every prediction by that amount. Because the guarantee is finite-sample and distribution-free under exchangeability, no retraining and no white-box access are needed. Across three benchmarks and five grounders, realized coverage tracks the requested target, and the calibration step exposes grounder-specific failure modes that point metrics hide.

What carries the argument

The central object is the temporal nonconformity score: for an interval grounder, $s = \max\big((\hat s - s^*)/\ell, \ (e^* - \hat e)/\ell\big)$, the larger of the two per-side overshoots normalized by predicted length; for a relevance-signal grounder, $s = -\min_{t \in I^*} f_X(t)$. These scores are fed through split-conformal calibration: $\hat q$ is the $\lceil (n+1)(1-\alpha)\rceil$-th smallest calibration score, and the output region is the nested family $\mathcal{C}_{\hat q}(X)$ (interval widened by $\hat q\ell$ per side, or the super-level set $\{t: f_X(t) \ge -\hat q\}$). The theory in Theorem 5 shows that within this one-parameter family the calibrated region is the shortest one with marginal coverage at least $1-\alpha$, and Proposition 8 bounds the coverage loss by a total-variation defect when exchangeability fails.

What would settle it

Take a long video, use early clips as calibration and later clips as test for the same query type, and run Algorithm 1. Because the clips are temporally dependent, exchangeability fails; if the measured coverage falls below $1-\alpha$ by more than the total-variation defect $\Delta$ computed from Proposition 8, the bound is violated.

Watch

Extended reading notes

Core claim

The central discovery is that the object of uncertainty quantification for temporal grounding can be the temporal region itself, not the point interval. Given any grounder, define the score of a query–video pair as the smallest widening of the predicted interval that makes it contain the true moment (or, for a relevance signal, the negative of its minimum over the true moment). Calibrating the finite-sample-corrected quantile of this score on a held-out set and widening each test prediction by that quantile yields a region that contains the true interval with probability at least $1-\alpha$ (Proposition 4). The paper also proves the calibrated region is asymptotically the shortest in its one-parameter family, gives conditions under which coverage holds within event-length strata, and bounds the coverage loss when moments from one video break exchangeability. Empirically, the wrapper delivers the requested coverage on nearly every configuration, and the calibrated boundary parameters diagnose each grounder's error structure.

Load-bearing premise

The guarantee collapses if the calibration and test query–video pairs are not exchangeable, and the paper's own experiments show this happens in practice: cross-dataset transfer can lose up to 18.8 points of coverage, and a VLM's within-dataset subgroups can be non-exchangeable after conditioning.

Editorial extensions

If this is right

  • Any existing video temporal grounder can be upgraded to a reliable region predictor with only a labeled calibration set and a single sort, making uncertainty quantification accessible for black-box APIs.
  • The calibrated region is asymptotically the tightest possible within its score family, so the wrapper adds no unnecessary temporal slack.
  • Under a scale model where relative boundary error is independent of event length, the normalized score yields length-conditional coverage that a seconds-valued score does not provide.
  • When calibration and test distributions differ, the coverage loss is bounded by the total-variation defect, and duration-bucketed reweighting can restore much of it.
  • Calibration acts as a diagnostic: the per-boundary calibrated parameters reveal asymmetric error (for example, a VLM with error-free onsets and poorly localized offsets), information point metrics do not carry.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same wrapper could extend to other interval-valued outputs, such as audio event detection or spatio-temporal tube localization, where a single scalar quantile would certify simultaneous boundary containment.
  • The paper assumes one ground-truth interval per query, but its own motivation highlights annotator disagreement; a natural extension would certify coverage against a quantile of the annotation interval distribution rather than a single label.
  • The certified width itself is a per-query difficulty signal; the paper's downstream-utility analysis shows it can drive selective prediction, but the sign of the width-quality correlation is dataset-dependent, so deployments should check that correlation before using width as a deferral rule.
  • On small, clustered calibration pools like QVHighlights, the paper's own results suggest that group-conditional or Mondrian calibration should be the default even when marginal coverage looks acceptable, because a few undercovered strata can drag the marginal average below target.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. The paper proposes COVER, a post-hoc, model-agnostic conformal wrapper for video temporal grounding. For a fixed grounder g (trained localizer or black-box VLM), COVER computes a temporal nonconformity score on a held-out calibration set, takes the (1-alpha) empirical quantile qhat of that score, and widens each test prediction by qhat (in a score-dependent way) to produce a temporal region claimed to contain the true interval I* with probability at least 1-alpha under exchangeability. Two score families are introduced: an interval-widening score based on the maximum per-side overshoot normalized by predicted length, and a super-level-set score for grounders exposing a relevance signal. The paper also presents theory on region-length optimality (Theorem 5), length-conditional coverage under a scale model (Proposition 7), and a bound on coverage loss under non-exchangeability (Proposition 8). Experiments on Charades-STA, ActivityNet-Captions, and QVHighlights with five grounders show that calibrated coverage tracks the target for most pairs, while hand-picked margins produce coverage between 0.096 and 1.000, and an evidential grounder (SRAM) misses its nominal level by 14 points.

Significance. If the central claim holds, the paper makes a useful and practical contribution: a single scalar calibration step converts any interval-producing or relevance-signal-producing grounder into one with finite-sample, distribution-free marginal coverage, requiring no retraining or white-box access. The paper is unusually explicit about its assumptions and failure modes, and it reports several honest and informative negative results (e.g., the QVHighlights QD-DETR coverage gap, the positivity limitation in cross-dataset transfer, the SRAM shortfall). The main proof of marginal coverage (Proposition 4) is the standard split-conformal rank argument and is correct. The paper also provides public code, a clear algorithmic description, and experiments across multiple datasets and grounders. The two-parameter per-boundary variant, which is presented as carrying the same guarantee and is used in the headline efficiency comparison, is not justified as written; this is a load-bearing gap that affects the central claim's scope and the validity of Table A2/A3 comparisons.

major comments (2)
  1. [Sec. 4 (Per-boundary variant) and Sec. 7.2, Tables A2–A3] The per-boundary variant calibrates qhat_s and qhat_e independently at the same (1-alpha) level and then states in Sec. 7.2 that "Both members carry the same coverage guarantee." This is not justified: the events that the start side is covered and that the end side is covered each hold with probability at least 1-alpha, but their intersection (joint containment of the whole interval I*) holds with probability at least 1-2alpha by the union bound, and exactly (1-alpha)^2 in the independent case. No Bonferroni correction, alpha/2 split, or joint-rank correction appears in Sec. 4 or in the Appendix. Consequently the efficiency comparisons in Tables A2–A3 are not at matched coverage: the two-parameter variant can appear shorter precisely because it undercovers. The claim can be repaired by calibrating each side at 1-alpha/2 (or by a joint correction), but then qhat_s, qhat_e, and the length savings in Tables A2–A3 would change, so the empirical comparison must be rerun and the text revised.
  2. [Sec. 7.1, Sec. 7.3, Sec. 8] The paper's headline claim is a coverage guarantee for "any video temporal grounder," but the experiments document two real violations of Assumption 1: QD-DETR on QVHighlights realizes 0.879 at target 0.90 (Sec. 7.1), attributed to within-dataset subgroup non-exchangeability, and cross-dataset transfer loses up to 18.8 percentage points of coverage with only partial repair by reweighting (Sec. 7.3, Table A13). These are not proof errors, and the paper discusses them in Sec. 8, but they mean the guarantee holds only under the stated exchangeability assumption, which the empirical findings show fails in realistic settings. The abstract and contributions should be qualified accordingly, and the QVHighlights result should be presented as a documented boundary case of the method's validity rather than as a one-off deviation, so readers do not infer unconditional coverage for arbitrary deployment shifts.
minor comments (5)
  1. [Sec. 7.2] The sentence "Tables A2 and A2 in the Appendix compare the single-parameter score against the per-boundary two-parameter variant" contains a duplicated table number; it should refer to Tables A2 and A3.
  2. [Figure A1 caption] The caption contains the typo "Fhe same grounder" and should read "The same grounder."
  3. [Sec. 7.1] The phrase "under the k/(n+1) floor Proposition 12 predicts" is ambiguous; it should say "below the k/(n+1) floor predicted by Proposition 12," since the observed coverage 0.879 is lower than the nominal target.
  4. [Sec. 3 and Sec. 8] The paper motivates the task by annotator disagreement and states that ground truth is a distribution over intervals, yet the formal setup and all calibration use a single ground-truth interval I* per query–video pair. The coverage guarantee is therefore coverage of that reference label, not of the distribution of human annotations; the paper should state this explicitly to avoid over-interpreting the guarantee in the presence of annotation variability.
  5. [Sec. 6] The statement that Lighthouse caps usable video length and removes 34.5% of ActivityNet-Captions' original pool should be acknowledged as a potentially significant distributional alteration, since the resulting benchmark may not be representative of the full dataset population.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the coverage guarantee is the standard split-conformal exchangeability argument; the calibrated quantile is the method, not a fitted constant renamed as a prediction.

full rationale

The central Proposition 4 derives coverage from Assumptions 1–3 by the standard exchangeable-rank argument: the temporal nonconformity score s(X,I*) is defined as the smallest widening λ that makes the nested region contain the truth (Eqs. 1–2; Assumption 2), and q̂ is the finite-sample-corrected empirical quantile (Eq. 3). The containment equivalence I* ⊆ C_q̂(X) ⇔ s ≤ q̂ is the definition of the score, not an imported conclusion, and the probability bound follows from exchangeability of s_1,…,s_{n+1}, with q̂ learned from held-out calibration data before test. No parameter is fitted to a subset and then reported as a prediction of a closely related quantity: q̂ is exactly the method's only fitted quantity, and coverage is evaluated on disjoint test data. Proposition 7's length-conditional claim is stated under an explicit scale-model assumption (independence of the normalized score and true event length) and then tested directionally on held-out strata, so it is a falsifiable hypothesis rather than a circular derivation. Theorem 5's length identity and within-family length-optimality follow from monotonicity of region length in λ and the definition of the quantile, not from the data. The cited conformal facts [15,17,20,22] are standard external results, and there are no load-bearing self-citations. The per-boundary two-parameter variant's statement that 'both members carry the same coverage guarantee' is, as written, a correctness gap (separate 1−α quantiles give at best 1−2α joint coverage unless a union-bound or α/2 correction is added), but that is an unsupported mathematical assertion, not a circular reduction of a prediction to its fitted inputs. Accordingly no circular step is identified; the appropriate score is 0.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The core wrapper uses one calibrated scalar and introduces no physical entities or hidden constants. The main assumptions are the standard conformal triplet of exchangeability, nested family, and frozen pipeline, plus a scale model for the length-conditional claim. The per-boundary variant adds two calibrated scalars but lacks a proven joint-coverage guarantee.

free parameters (3)
  • calibration quantile qhat = data-dependent; e.g., 2.035 for QD-DETR norm on Charades at alpha=0.1 in Table A9
    Set as the ceil((n+1)(1-alpha))-th smallest nonconformity score; this is the method's only fitted scalar for the single-parameter wrapper and is not an ad hoc constant.
  • per-boundary quantiles qhat_start and qhat_end = e.g., Qwen QVHighlights sec at 1-alpha=0.90: 3.80 and 115.40 in Table A3
    Separately calibrated boundary thresholds in the two-parameter variant; their joint coverage guarantee is not proved in the paper.
  • hand-picked fixed margins lambda = 5, 10, 20, 40, 80, 160 seconds = six constants from Table 3
    Used only as an uncalibrated baseline to show that margins cannot be chosen by hand; not part of Cover itself.
assumptions (6)
  • domain assumption Assumption 1: Exchangeability of calibration and test pairs
    Calibration and test pairs are drawn exchangeably; this is needed for score rank uniformity. Violated by cross-dataset shift and within-video dependence, for which Proposition 8 quantifies the loss.
  • domain assumption Assumption 2: Nested region family with score equivalence I* subseteq C_lambda iff s <= lambda
    This makes the nonconformity score a valid threshold and makes the miss-loss monotone in lambda; it is satisfied by the interval-widening and super-level-set families.
  • domain assumption Assumption 3: Frozen pipeline, grounder and score fixed before calibration labels are seen
    Required so the scores remain exchangeable; tuning g or s on calibration labels would invalidate the guarantee.
  • ad hoc to paper Scale model: normalized score independent of true event length
    Assumed in Proposition 7 to prove length-conditional coverage; the paper tests the directional prediction but does not establish the assumption independently.
  • standard math i.i.d. calibration scores for Theorem 5(ii)
    Strengthens exchangeability to independence to obtain consistency and the sqrt(n) rate of qhat.
  • standard math Continuous score distribution for Theorem 5(iii)-(iv)
    Needed for unique quantiles and strict monotonicity of coverage and length in the family-optimality statements.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Conformal Coverage Guarantees for Any Video Temporal Grounder." pith.science (2026). https://pith.science/paper/HRUITLIW

@misc{pith2026260807434,
  author       = {Pith},
  title        = {Pith review of: Conformal Coverage Guarantees for Any Video Temporal Grounder},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HRUITLIW}},
  note         = {Machine review of arXiv:2608.07434}
}
abstract

Event boundaries in continuous video are ambiguous: re-annotate the same query-video pair and independent annotators mark moments that overlap by less than half on a large fraction of samples. The ground truth for video temporal grounding is therefore a distribution over intervals, yet every grounder returns a single interval with no statement of reliability, so at deployment a wrong interval is indistinguishable from a right one. COVER changes the output object: a post-hoc, model-agnostic wrapper that turns any grounder, a trained localizer or a black-box video--language model, into one that emits a temporal region containing the true moment with probability at least $1-\alpha$, by calibrating the quantile of a temporal nonconformity score on held-out labels and widening the base prediction by that amount. The guarantee is finite-sample and distribution-free under exchangeability, and requires neither retraining nor white-box access. We give two score families, a two-sided boundary-widening score for grounders that emit an interval and a super-level-set score for grounders that emit a relevance signal, and develop theory specific to grounding that bounds how large the certified region becomes, when coverage survives conditioning on event length, and how it degrades when moments from one video break exchangeability. Across three benchmarks and five grounders, realized coverage tracks the target, and calibration exposes what point metrics hide.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 32 canonical work pages

  1. [1]

    Uncovering hidden challenges in query-based video moment retrieval.arXiv preprint arXiv:2009.00325, 2020

    Mayu Otani, Yuta Nakashima, Esa Rahtu, and Janne Heikkilä. Uncovering hidden challenges in query-based video moment retrieval.arXiv preprint arXiv:2009.00325, 2020

    arXiv 2009

  2. [2]

    Towards Diverse Temporal Grounding under Single Positive Labels

    Hao Zhou, Chongyang Zhang, Yanjun Chen, and Chuanping Hu. Towards diverse temporal grounding under single positive labels.arXiv preprint arXiv:2303.06545, 2023

    work page Pith review arXiv 2023

  3. [3]

    Temporal sentence grounding in videos: A survey and future directions.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023

    Hao Zhang, Aixin Sun, Wei Jing, and Joey Tianyi Zhou. Temporal sentence grounding in videos: A survey and future directions.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023

    work page 2023

  4. [4]

    TALL: Temporal activity localization via language query

    Jiyang Gao, Chen Sun, Zhenheng Yang, and Ram Nevatia. TALL: Temporal activity localization via language query. InProceedings of the IEEE International Conference on Computer Vision (ICCV), 2017

    work page 2017

  5. [5]

    Dense-captioning events in videos

    Ranjay Krishna, Kenji Hata, Frederic Ren, Li Fei-Fei, and Juan Carlos Niebles. Dense-captioning events in videos. InProceedings of the IEEE International Conference on Computer Vision (ICCV), 2017

    work page 2017

  6. [6]

    Detecting moments and highlights in videos via natural language queries

    Jie Lei, Tamara L Berg, and Mohit Bansal. Detecting moments and highlights in videos via natural language queries. InAdvances in Neural Information Processing Systems (NeurIPS), 2021

    work page 2021

  7. [7]

    Query-dependent video representation for moment retrieval and highlight detection

    WonJun Moon, Sangeek Hyun, SangUk Park, Dongchan Park, and Jae-Pil Heo. Query-dependent video representation for moment retrieval and highlight detection. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023

    work page 2023

  8. [8]

    BAM-DETR: Boundary-aligned moment detection transformer for temporal sentence grounding in videos

    Pilhyeon Lee and Hyeran Byun. BAM-DETR: Boundary-aligned moment detection transformer for temporal sentence grounding in videos. InEuropean Conference on Computer Vision (ECCV), 2024

    work page 2024

Show all 44 references
  1. [9]

    UniVTG: Towards unified video-language temporal grounding

    Kevin Qinghong Lin, Pengchuan Zhang, Joya Chen, Shraman Pramanick, Difei Gao, Alex Jinpeng Wang, Rui Yan, and Mike Zheng Shou. UniVTG: Towards unified video-language temporal grounding. InProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2023

    2023

  2. [10]

    VTimeLLM: Empower LLM to grasp video moments.arXiv preprint arXiv:2311.18445, 2023

    Bin Huang, Xin Wang, Hong Chen, Zihan Song, and Wenwu Zhu. VTimeLLM: Empower LLM to grasp video moments.arXiv preprint arXiv:2311.18445, 2023

    2023 arXiv

  3. [11]

    Training-free video temporal 14 CONFORMAL COVERAGE GUARANTEES FOR ANY VIDEO TEMPORAL GROUNDER Kurban Intelligence Lab grounding using large-scale pre-trained models

    Minghang Zheng, Xinhao Cai, Qingchao Chen, Yuxin Peng, and Yang Liu. Training-free video temporal 14 CONFORMAL COVERAGE GUARANTEES FOR ANY VIDEO TEMPORAL GROUNDER Kurban Intelligence Lab grounding using large-scale pre-trained models. InEuropean Conference on Computer Vision (...

    2024

  4. [12]

    Foresee-to-ground: From predictive temporal perception to evidence-driven reasoning for video tem- poral grounding.arXiv preprint arXiv:2605.21973, 2026

    Zelin Zheng, Xinyan Liu, Ruixin Li, Antoni B Chan, Guorong Li, Qingming Huang, and Laiyun Qing. Foresee-to-ground: From predictive temporal perception to evidence-driven reasoning for video tem- poral grounding.arXiv preprint arXiv:2605.21973, 2026

    2026 arXiv

  5. [13]

    Beyond uncertainty: Evidential deep learning for robust video temporal grounding.arXiv preprint arXiv:2408.16272, 2024

    Kaijing Ma, Haojian Huang, Jin Chen, Haodong Chen, Pengliang Ji, Xianghao Zang, Han Fang, Chao Ban, Hao Sun, Mulin Chen, et al. Beyond uncertainty: Evidential deep learning for robust video temporal grounding.arXiv preprint arXiv:2408.16272, 2024

    2024 arXiv

  6. [14]

    Adaptive evidential learning for temporal-semantic robustness in moment retrieval

    Haojian Huang, Kaijing Ma, Jin Chen, Haodong Chen, Zhou Wu, Xianghao Zang, Han Fang, Chao Ban, Hao Sun, Mulin Chen, et al. Adaptive evidential learning for temporal-semantic robustness in moment retrieval. InProceedings of the AAAI Conference on Artificial Intelligence, volume...

    2026

  7. [15]

    Springer, 2005

    Vladimir Vovk, Alexander Gammerman, and Glenn Shafer.Algorithmic Learning in a Random World. Springer, 2005

    2005

  8. [16]

    A gentle introduction to conformal prediction and distribution-free uncertainty quantification.Foundations and Trends in Machine Learning, 16(4):494–591, 2023

    Anastasios N Angelopoulos and Stephen Bates. A gentle introduction to conformal prediction and distribution-free uncertainty quantification.Foundations and Trends in Machine Learning, 16(4):494–591, 2023

    2023

  9. [17]

    Distribution-free predictive inference for regression.Journal of the American Statistical Association, 113(523):1094–1111, 2018

    Jing Lei, Max G’Sell, Alessandro Rinaldo, Ryan J Tibshirani, and Larry Wasserman. Distribution-free predictive inference for regression.Journal of the American Statistical Association, 113(523):1094–1111, 2018

    2018

  10. [18]

    Theoretical foundations of conformal prediction.arXiv preprint arXiv:2411.11824, 2024

    Anastasios N Angelopoulos, Rina Foygel Barber, and Stephen Bates. Theoretical foundations of conformal prediction.arXiv preprint arXiv:2411.11824, 2024

    2024 arXiv

  11. [19]

    Uncertainty sets for image classifiers using conformal prediction

    Anastasios N Angelopoulos, Stephen Bates, Jitendra Malik, and Michael I Jordan. Uncertainty sets for image classifiers using conformal prediction. InInternational Conference on Learning Representations (ICLR), 2021

    2021

  12. [20]

    Conformalized quantile regression

    Yaniv Romano, Evan Patterson, and Emmanuel J Candès. Conformalized quantile regression. In Advances in Neural Information Processing Systems (NeurIPS), 2019

    2019

  13. [21]

    Copula-based conformal prediction for object detection: A more efficient approach

    Bruce Cyusa Mukama, Soundouss Messoudi, Sylvain Rousseau, and Sébastien Destercke. Copula-based conformal prediction for object detection: A more efficient approach. InSymposium on Conformal and Probabilistic Prediction with Applications (COPA), PMLR, 2024

    2024

  14. [22]

    Distribution- free, risk-controlling prediction sets.Journal of the ACM, 68(6):1–34, 2021

    Stephen Bates, Anastasios Angelopoulos, Lihua Lei, Jitendra Malik, and Michael I Jordan. Distribution- free, risk-controlling prediction sets.Journal of the ACM, 68(6):1–34, 2021

    2021

  15. [23]

    Conformal object detection by sequential risk control.arXiv preprint arXiv:2505.24038, 2025

    Léo Andéol, Luca Mossina, Adrien Mazoyer, and Sébastien Gerchinovitz. Conformal object detection by sequential risk control.arXiv preprint arXiv:2505.24038, 2025

    2025

  16. [24]

    Adaptive bound- ing box uncertainties via two-step conformal prediction.arXiv preprint arXiv:2403.07263, 2024

    Alexander Timans, Christoph-Nikolas Straehle, Kaspar Sakmann, and Eric Nalisnick. Adaptive bound- ing box uncertainties via two-step conformal prediction.arXiv preprint arXiv:2403.07263, 2024

    2024 arXiv

  17. [25]

    Conformal prediction sets for instance segmentation.arXiv preprint arXiv:2602.10045, 2026

    Kerri Lu, Dan M Kluger, Stephen Bates, and Sherrie Wang. Conformal prediction sets for instance segmentation.arXiv preprint arXiv:2602.10045, 2026

    2026 arXiv

  18. [26]

    Explicit abstention knobs for predictable reliability in video question answering.arXiv preprint arXiv:2601.00138, 2026

    Jorge Ortiz. Explicit abstention knobs for predictable reliability in video question answering.arXiv preprint arXiv:2601.00138, 2026

    2026

  19. [27]

    Conformal risk control

    Anastasios N Angelopoulos, Stephen Bates, Adam Fisch, Lihua Lei, and Tal Schuster. Conformal risk control. InInternational Conference on Learning Representations (ICLR), 2024

    2024

  20. [28]

    Video activity localisation with uncertainties in temporal boundary

    Jiabo Huang, Yang Liu, Shaogang Gong, and Hailin Jin. Video activity localisation with uncertainties in temporal boundary. InEuropean Conference on Computer Vision (ECCV), 2022. 15 CONFORMAL COVERAGE GUARANTEES FOR ANY VIDEO TEMPORAL GROUNDER Kurban Intelligence Lab

    2022

  21. [29]

    Learn then test: Calibrating predictive algorithms to achieve risk control.arXiv preprint arXiv:2110.01052, 2021

    Anastasios N Angelopoulos, Stephen Bates, Emmanuel J Candès, Michael I Jordan, and Lihua Lei. Learn then test: Calibrating predictive algorithms to achieve risk control.arXiv preprint arXiv:2110.01052, 2021

    2021 arXiv

  22. [30]

    The limits of distribution-free conditional predictive inference.Information and Inference, 10(2):455–482, 2021

    Rina Foygel Barber, Emmanuel J Candès, Aaditya Ramdas, and Ryan J Tibshirani. The limits of distribution-free conditional predictive inference.Information and Inference, 10(2):455–482, 2021

    2021

  23. [31]

    Conditional validity of inductive conformal predictors

    Vladimir Vovk. Conditional validity of inductive conformal predictors. InAsian Conference on Machine Learning (ACML), 2012

    2012

  24. [32]

    Conformal prediction with conditional guarantees

    Isaac Gibbs, John J Cherian, and Emmanuel J Candès. Conformal prediction with conditional guarantees. arXiv preprint arXiv:2305.12616, 2024

    2024 arXiv

  25. [33]

    Conformal prediction under covariate shift

    Ryan J Tibshirani, Rina Foygel Barber, Emmanuel J Candès, and Aaditya Ramdas. Conformal prediction under covariate shift. InAdvances in Neural Information Processing Systems (NeurIPS), 2019

    2019

  26. [34]

    Conformal prediction beyond exchangeability.Annals of Statistics, 51(2):816–845, 2023

    Rina Foygel Barber, Emmanuel J Candès, Aaditya Ramdas, and Ryan J Tibshirani. Conformal prediction beyond exchangeability.Annals of Statistics, 51(2):816–845, 2023

    2023

  27. [35]

    Adaptive conformal inference under distribution shift

    Isaac Gibbs and Emmanuel Candès. Adaptive conformal inference under distribution shift. InAdvances in Neural Information Processing Systems (NeurIPS), 2021

    2021

  28. [36]

    Confident object detection via conformal prediction and conformal risk control: an application to railway signaling

    Léo Andéol, Thomas Fel, Florence De Grancey, and Luca Mossina. Confident object detection via conformal prediction and conformal risk control: an application to railway signaling. InSymposium on Conformal and Probabilistic Prediction with Applications (COPA), PMLR, 2023

    2023

  29. [37]

    Conformal time-series forecasting

    Kamile Stankeviciute, Ahmed M Alaa, and Mihaela van der Schaar. Conformal time-series forecasting. InAdvances in Neural Information Processing Systems (NeurIPS), 2021

    2021

  30. [38]

    Lighthouse: A user-friendly library for reproducible video moment retrieval and highlight detection

    Taichi Nishimura, Shota Nakada, Hokuto Munakata, and Tatsuya Komatsu. Lighthouse: A user-friendly library for reproducible video moment retrieval and highlight detection. InProceedings of The 2024 Conference on Empirical Methods in Natural Language Processing: System Demonstra...

    2024

  31. [39]

    Qwen2.5-VL technical report.arXiv preprint arXiv:2502.13923, 2025

    Shuai Bai et al. Qwen2.5-VL technical report.arXiv preprint arXiv:2502.13923, 2025

    2025 arXiv

  32. [40]

    Non-exchangeable confor- mal risk control.arXiv preprint arXiv:2310.01262, 2023

    António Farinhas, Chrysoula Zerva, Dennis Ulmer, and André F T Martins. Non-exchangeable confor- mal risk control.arXiv preprint arXiv:2310.01262, 2023

    2023 arXiv

  33. [41]

    Oxford University Press, 2013

    Stéphane Boucheron, Gábor Lugosi, and Pascal Massart.Concentration Inequalities: A Nonasymptotic Theory of Independence. Oxford University Press, 2013

    2013

  34. [42]

    Tr-detr: Task-reciprocal transformer for joint moment retrieval and highlight detection

    Hao Sun, Mingyao Zhou, Wenjing Chen, and Wei Xie. Tr-detr: Task-reciprocal transformer for joint moment retrieval and highlight detection. InProceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 4998–5007, 2024

    2024

  35. [43]

    Internvl3

    Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, Long Cui, Xingguang Wei, Zhaoyang Liu, Linglin Jing, Shenglong Ye, Jie Shao, et al. Internvl3. 5: Advancing open-source multimodal models in versatility, reasoning, and efficiency.arXiv preprint arXiv:2508.18265, 2025

    2025 arXiv

  36. [44]

    Vast.ai, 2026

    Vast.ai. Vast.ai, 2026. URLhttps://vast.ai/. 16 CONFORMAL COVERAGE GUARANTEES FOR ANY VIDEO TEMPORAL GROUNDER Kurban Intelligence Lab APPENDIX A Standard Guarantees These results are split-conformal and risk-control facts [15, 17, 22] instantiated with the temporal score. We s...

    2026

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.