# TOD

Source: [TOD](<https://www.sota2.com/research/dataset/tod>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Benchmarks

Page 1 of 2. This is an index page, not the full collection.

- [Next index page](<https://www.sota2.com/research/dataset/tod.md?page=2>): Follow this link to discover more pages.

- [Object Detection on AI-TOD (test)](<https://www.sota2.com/research/sota/object-detection-on-ai-tod-test.md>): Object Detection; AI-TOD (test).

- [Jailbreak Defense on AutoDAN](<https://www.sota2.com/research/sota/jailbreak-defense-on-autodan.md>): Jailbreak Defense; AutoDAN.

- [6D Object Pose Estimation on BOP 7 Core Datasets (test)](<https://www.sota2.com/research/sota/6d-object-pose-estimation-on-bop-7-core-datasets-test.md>): 6D Object Pose Estimation; BOP 7 core datasets: LM-O, T-LESS, TUD-L, IC-BIN, ITODD, HB, YCB-V 82 (test).

- [Jailbreak Attack on AutoDAN](<https://www.sota2.com/research/sota/jailbreak-attack-on-autodan.md>): Jailbreak Attack; AutoDAN.

- [Jailbreak Detection on AutoDAN](<https://www.sota2.com/research/sota/jailbreak-detection-on-autodan.md>): Jailbreak Detection; AutoDAN.

- [3D Tracking on PointOdyssey 102 (test)](<https://www.sota2.com/research/sota/3d-tracking-on-pointodyssey-102-test.md>): 3D tracking; PointOdyssey 102 (test).

- [Binary decision on TODAY](<https://www.sota2.com/research/sota/binary-decision-on-today.md>): Binary decision; TODAY.

- [Task-oriented dialogue on FnCTOD](<https://www.sota2.com/research/sota/task-oriented-dialogue-on-fnctod.md>): Task-oriented dialogue; FnCTOD.

- [Jailbreak Defense on Jailbreak Attack Suite (GCG, AutoDAN, DeepInc., Prefill, Int.Laund.)](<https://www.sota2.com/research/sota/jailbreak-defense-on-jailbreak-attack-suite-gcg-autodan-deepinc-prefill-int-laund.md>): Jailbreak Defense; Jailbreak Attack Suite GCG, AutoDAN, DeepInc., Prefill, Int.Laund..

- [Time-Series Anomaly Detection on TODS](<https://www.sota2.com/research/sota/time-series-anomaly-detection-on-tods.md>): Time-Series Anomaly Detection; TODS.

- [Object Detection on AI-TOD 1.0 (test)](<https://www.sota2.com/research/sota/object-detection-on-ai-tod-1-0-test.md>): Object Detection; AI-TOD 1.0 (test).

- [Jailbreak Defense on GPTFuzz, TAP, GCG, AutoDAN, and Template](<https://www.sota2.com/research/sota/jailbreak-defense-on-gptfuzz-tap-gcg-autodan-and-template.md>): Jailbreak Defense; Jailbreak Attack Benchmarks (GPTFuzz, TAP, GCG, AutoDAN, Template).

- [Video Depth Estimation on PointOdyssey (val)](<https://www.sota2.com/research/sota/video-depth-estimation-on-pointodyssey-val.md>): Video depth estimation; PointOdyssey (val).

- [Object Detection on DTOD](<https://www.sota2.com/research/sota/object-detection-on-dtod.md>): Object Detection; DTOD.

- [Reference-grounded News Verification on USA Today](<https://www.sota2.com/research/sota/reference-grounded-news-verification-on-usa-today.md>): Reference-grounded News Verification; USA Today.

- [Jailbreak Defense on AdvBench AutoDAN](<https://www.sota2.com/research/sota/jailbreak-defense-on-advbench-autodan.md>): Jailbreak Defense; AdvBench AutoDAN.

- [Adversarial Detection on AutoDAN](<https://www.sota2.com/research/sota/adversarial-detection-on-autodan.md>): Adversarial Detection; AutoDAN.

- [Adversarial Robustness on AutoDAN](<https://www.sota2.com/research/sota/adversarial-robustness-on-autodan.md>): Adversarial Robustness; AutoDAN.

- [Binary temporal relation classification on TRACIE, MATRES, and TODAY (test)](<https://www.sota2.com/research/sota/binary-temporal-relation-classification-on-tracie-matres-and-today-test.md>): Binary temporal relation classification; TRACIE, MATRES, and TODAY (test).

- [Safety Evaluation on AutoDAN](<https://www.sota2.com/research/sota/safety-evaluation-on-autodan.md>): Safety Evaluation; AutoDAN.

- [Defense Robustness on AutoDAN Attack](<https://www.sota2.com/research/sota/defense-robustness-on-autodan-attack.md>): Defense Robustness; AutoDAN Attack.

- [Jailbreak Detection on AutoDAN (ASR/PGR)](<https://www.sota2.com/research/sota/jailbreak-detection-on-autodan-asr-pgr.md>): Jailbreak Detection; AutoDAN.

- [Jailbreak Defense on AdvBench (AutoDAN Attack)](<https://www.sota2.com/research/sota/jailbreak-defense-on-advbench-autodan-attack.md>): Jailbreak Defense; AdvBench AutoDAN attack.

- [Video Denoising on Bosch Autodome IP 5000 IR Windmill sequence](<https://www.sota2.com/research/sota/video-denoising-on-bosch-autodome-ip-5000-ir-windmill-sequence.md>): Video Denoising; Bosch Autodome IP 5000 IR Windmill sequence.

- [Object Detection on CytoDArk0](<https://www.sota2.com/research/sota/object-detection-on-cytodark0.md>): Object Detection; CytoDArk0.

- [Cell Detection on CytoDArk0 (test)](<https://www.sota2.com/research/sota/cell-detection-on-cytodark0-test.md>): Cell Detection; CytoDArk0 (test).

- [Jailbreak Defense on AutoDAN AdvE](<https://www.sota2.com/research/sota/jailbreak-defense-on-autodan-adve.md>): Jailbreak Defense; AutoDAN AdvE.

- [Jailbreak Defense on Common Jailbreak Attacks Suite (Qwen3.5-9B)](<https://www.sota2.com/research/sota/jailbreak-defense-on-common-jailbreak-attacks-suite-qwen3-5-9b.md>): Jailbreak Defense; Common Jailbreak Attacks Suite (Manual (IJP), GCG, AutoDAN, DrAttack, MultiJail, ActorAttack) on Qwen3.5-9B (test).

- [Jailbreak Defense on AutoDAN (ASR, PGR)](<https://www.sota2.com/research/sota/jailbreak-defense-on-autodan-asr-pgr.md>): Jailbreak Defense; AutoDAN.

- [6D Object Pose Estimation on BOP (T-LESS, ITODD, YCB-V, LM-O) Challenge (test)](<https://www.sota2.com/research/sota/6d-object-pose-estimation-on-bop-t-less-itodd-ycb-v-lm-o-challenge-test.md>): 6D Object Pose Estimation; BOP (T-LESS, ITODD, YCB-V, LM-O) Challenge (test).

- [Point Tracking on PointOdyssey (test)](<https://www.sota2.com/research/sota/point-tracking-on-pointodyssey-test.md>): Point Tracking; PointOdyssey (test).

- [3D Reconstruction on CMU Panoptic-Toddler (test)](<https://www.sota2.com/research/sota/3d-reconstruction-on-cmu-panoptic-toddler-test.md>): 3D Reconstruction; CMU Panoptic-Toddler (test).

- [Skin condition differential diagnosis on TODIV cohort](<https://www.sota2.com/research/sota/skin-condition-differential-diagnosis-on-todiv-cohort.md>): Skin condition differential diagnosis; TODIV cohort.

- [Adversarial Attack Defense on Safety Evaluation Set GCG, AutoDAN, DeepInception, Prefilling, Intent Laundering n=200 (full public set)](<https://www.sota2.com/research/sota/adversarial-attack-defense-on-safety-evaluation-set-gcg-autodan-deepinception-prefilling-intent-laundering-n-200-full-public-set.md>): Adversarial Attack Defense; Safety Evaluation Set GCG, AutoDAN, DeepInception, Prefilling, Intent Laundering n=200 (full public set).

- [Jailbreak Attack on AutoDAN T-R](<https://www.sota2.com/research/sota/jailbreak-attack-on-autodan-t-r.md>): Jailbreak Attack; AutoDAN T-R.

- [Object Detection on AI-TOD v2 (val)](<https://www.sota2.com/research/sota/object-detection-on-ai-tod-v2-val.md>): Object Detection; AI-TOD v2 (val).

- [Sparse Point Tracking on PointOdyssey (test)](<https://www.sota2.com/research/sota/sparse-point-tracking-on-pointodyssey-test.md>): Sparse Point Tracking; PointOdyssey (PO) (test).

- [End-to-End Dialogue Modeling on ToDs (test)](<https://www.sota2.com/research/sota/end-to-end-dialogue-modeling-on-tods-test.md>): End-to-End Dialogue Modeling; ToDs (test).

- [Natural Language Generation on ToDs benchmark GPT-2 backbone (test)](<https://www.sota2.com/research/sota/natural-language-generation-on-tods-benchmark-gpt-2-backbone-test.md>): Natural Language Generation; ToDs benchmark GPT-2 backbone (test).

- [Dialogue State Tracking on ToDs benchmark GPT-2 backbone (test)](<https://www.sota2.com/research/sota/dialogue-state-tracking-on-tods-benchmark-gpt-2-backbone-test.md>): Dialogue State Tracking; ToDs benchmark GPT-2 backbone (test).

- [Intent Classification on ToDs benchmark GPT-2 backbone (test)](<https://www.sota2.com/research/sota/intent-classification-on-tods-benchmark-gpt-2-backbone-test.md>): Intent Classification; ToDs benchmark GPT-2 backbone (test).

- [Reasoning/Planning on Today](<https://www.sota2.com/research/sota/reasoning-planning-on-today.md>): Reasoning/Planning; Today.

- [Object Detection on RTOD (test)](<https://www.sota2.com/research/sota/object-detection-on-rtod-test.md>): Object Detection; RTOD (test).

- [Multi-camera point tracking on PointOdyssey 1.0 (test)](<https://www.sota2.com/research/sota/multi-camera-point-tracking-on-pointodyssey-1-0-test.md>): Multi-camera point tracking; PointOdyssey 1.0 (test).

- [Video Denoising on Bosch Autodome IP 5000 IR (train)](<https://www.sota2.com/research/sota/video-denoising-on-bosch-autodome-ip-5000-ir-train.md>): Video Denoising; Bosch Autodome IP 5000 IR (train).

- [Defending against Gradient-based Attacks on Llama3 AutoDAN Attack (test)](<https://www.sota2.com/research/sota/defending-against-gradient-based-attacks-on-llama3-autodan-attack-test.md>): Defending against gradient-based attacks; Llama3 AutoDAN Attack (test).

- [Jailbreak Robustness on AdvBench AutoDAN-Turbo (original)](<https://www.sota2.com/research/sota/jailbreak-robustness-on-advbench-autodan-turbo-original.md>): Jailbreak Robustness; AdvBench AutoDAN-Turbo (original).

- [Two-view tracking on PointOdyssey (test)](<https://www.sota2.com/research/sota/two-view-tracking-on-pointodyssey-test.md>): Two-view tracking; PointOdyssey (test).

- [Jailbreak Detection on JailbreakBench AutoDAN-T](<https://www.sota2.com/research/sota/jailbreak-detection-on-jailbreakbench-autodan-t.md>): Jailbreak Detection; JailbreakBench AutoDAN-T.

- [Structural Reasoning on TODAY](<https://www.sota2.com/research/sota/structural-reasoning-on-today.md>): Structural Reasoning; TODAY.

- [Orientation Estimation on ITODD (ViVo)](<https://www.sota2.com/research/sota/orientation-estimation-on-itodd-vivo.md>): Orientation Estimation; ITODD (ViVo).

- [Orientation Estimation on ITODD SiSo](<https://www.sota2.com/research/sota/orientation-estimation-on-itodd-siso.md>): Orientation Estimation; ITODD SiSo.

- [Status Tracking on ATOD Complex (test)](<https://www.sota2.com/research/sota/status-tracking-on-atod-complex-test.md>): Status Tracking; ATOD Complex (test).

- [Goal Detection on ATOD Complex (test)](<https://www.sota2.com/research/sota/goal-detection-on-atod-complex-test.md>): Goal Detection; ATOD Complex (test).

- [Status Tracking on ATOD Medium (test)](<https://www.sota2.com/research/sota/status-tracking-on-atod-medium-test.md>): Status Tracking; ATOD Medium (test).

- [Goal Detection on ATOD Medium (test)](<https://www.sota2.com/research/sota/goal-detection-on-atod-medium-test.md>): Goal Detection; ATOD Medium (test).

- [Attack Success Rate Correlation Analysis on GCG/AutoDAN sparse-transfer diagnostic](<https://www.sota2.com/research/sota/attack-success-rate-correlation-analysis-on-gcg-autodan-sparse-transfer-diagnostic.md>): Attack Success Rate Correlation Analysis; GCG/AutoDAN sparse-transfer diagnostic.

- [Jailbreak Robustness on AutoDAN Adv single-turn attack](<https://www.sota2.com/research/sota/jailbreak-robustness-on-autodan-adv-single-turn-attack.md>): Jailbreak Robustness; AutoDAN Adv single-turn attack.

- [Jailbreak Robustness on AutoDAN Harm single-turn attack](<https://www.sota2.com/research/sota/jailbreak-robustness-on-autodan-harm-single-turn-attack.md>): Jailbreak Robustness; AutoDAN Harm single-turn attack.

- [Safety Defense Evaluation on Jailbreak Attacks (GCG, AutoDAN, PAIR, DeepInc., SAP30, SIJ)](<https://www.sota2.com/research/sota/safety-defense-evaluation-on-jailbreak-attacks-gcg-autodan-pair-deepinc-sap30-sij.md>): Safety Defense Evaluation; Jailbreak Attacks GCG, AutoDAN, PAIR, DeepInc., SAP30, SIJ.

- [Explanation Faithfulness on AutoDAN (test)](<https://www.sota2.com/research/sota/explanation-faithfulness-on-autodan-test.md>): Explanation Faithfulness; AutoDAN (test).

- [Anomaly Detection on TODS univariate](<https://www.sota2.com/research/sota/anomaly-detection-on-tods-univariate.md>): Anomaly Detection; TODS univariate.

- [4D reconstruction on PointOdyssey](<https://www.sota2.com/research/sota/4d-reconstruction-on-pointodyssey.md>): 4D reconstruction; PointOdyssey.

- [Predictive Validity Assessment on Mental Health Task-Oriented Dialogue (TOD) 160K (test)](<https://www.sota2.com/research/sota/predictive-validity-assessment-on-mental-health-task-oriented-dialogue-tod-160k-test.md>): Predictive Validity Assessment; Mental Health Task-Oriented Dialogue (TOD) 160K (test).

- [Safety Evaluation on AutoDAN (test)](<https://www.sota2.com/research/sota/safety-evaluation-on-autodan-test.md>): Safety Evaluation; AutoDAN (test).

- [Manipulation Localization on MDSM USA Today](<https://www.sota2.com/research/sota/manipulation-localization-on-mdsm-usa-today.md>): Manipulation Localization; MDSM USA Today.

- [Manipulation Detection on MDSM USA Today](<https://www.sota2.com/research/sota/manipulation-detection-on-mdsm-usa-today.md>): Manipulation Detection; MDSM USA Today.

- [4D Tracking on PointOdyssey & Dynamic Replica](<https://www.sota2.com/research/sota/4d-tracking-on-pointodyssey-and-dynamic-replica.md>): 4D Tracking; PointOdyssey & Dynamic Replica.

- [Language Modeling on Todd style corpus](<https://www.sota2.com/research/sota/language-modeling-on-todd-style-corpus.md>): Language Modeling; Todd style corpus.

- [Tiny Object Detection on AI-TOD v2 (test)](<https://www.sota2.com/research/sota/tiny-object-detection-on-ai-tod-v2-test.md>): Tiny Object Detection; AI-TOD v2 (test).

- [Confidence Estimation on MediTOD](<https://www.sota2.com/research/sota/confidence-estimation-on-meditod.md>): Confidence Estimation; MediTOD.

- [Dialogue Summarization on TODSum (test)](<https://www.sota2.com/research/sota/dialogue-summarization-on-todsum-test.md>): Dialogue Summarization; TODSum (test).

- [Dialogue Summarization on TODSum](<https://www.sota2.com/research/sota/dialogue-summarization-on-todsum.md>): Dialogue Summarization; TODSum.

- [Attack Success Rate on AutoDAN](<https://www.sota2.com/research/sota/attack-success-rate-on-autodan.md>): Safety Evaluation; AutoDAN.

- [Jailbreak Defense on Qwen2.5-7B (Adaptive AutoDAN-T Attack)](<https://www.sota2.com/research/sota/jailbreak-defense-on-qwen2-5-7b-adaptive-autodan-t-attack.md>): Jailbreak Defense; Qwen2.5-7B Adaptive AutoDAN-T attack.

- [Jailbreak Defense on LLaMA2-7B (Adaptive AutoDAN-T Attack)](<https://www.sota2.com/research/sota/jailbreak-defense-on-llama2-7b-adaptive-autodan-t-attack.md>): Jailbreak Defense; LLaMA2-7B Adaptive AutoDAN-T attack.

- [Jailbreak Defense on Vicuna-13B (Adaptive AutoDAN-T Attack)](<https://www.sota2.com/research/sota/jailbreak-defense-on-vicuna-13b-adaptive-autodan-t-attack.md>): Jailbreak Defense; Vicuna-13B Adaptive AutoDAN-T attack.

- [3D Reconstruction on CMU Panoptic Toddler](<https://www.sota2.com/research/sota/3d-reconstruction-on-cmu-panoptic-toddler.md>): 3D Reconstruction; CMU Panoptic Toddler.

- [6D Object Pose Estimation on TOD](<https://www.sota2.com/research/sota/6d-object-pose-estimation-on-tod.md>): 6D Object Pose Estimation; TOD.

- [Tiny Object Detection on AI-TOD (test)](<https://www.sota2.com/research/sota/tiny-object-detection-on-ai-tod-test.md>): Tiny Object Detection; AI-TOD (test).
