Research

Evaluates protein language models and geometric deep learning architectures on five realistic downstream biological tasks, including binding affinity prediction, functional annotation, mutation effects, cleavage site detection, and PROTAC interaction modeling. It probes how pretraining objectives, structural information integration, and domain-specific inductive biases affect generalization on limited biological data. Use when the user wants to benchmark on Protap Benchmark, or asks about eva...

Evaluates conversational agents on dialogue safety classification, rule-of-thumb generation, and prosocial response generation. It probes the model's ability to identify unsafe content, generate socially informed guidelines, and produce safe, engaging, and respectful dialogue responses. Use when the user wants to benchmark on PROSOCIALDIALOG, or asks about evaluating this task. Reports accuracy, BLEU-4.

Evaluates a model's ability to decompose sentences into atomic semantic units (propositional segmentation) and determine entailment relationships between text spans. It probes fine-grained compositional semantic alignment and partial entailment recognition beyond sentence-level NLI. Use when the user wants to benchmark on PropSegmEnt, or asks about evaluating this task. Reports Precision/Recall/F1w (macro-averaged).

Evaluates a two-stage malware detection framework that uses a fast ML classifier for initial triage and a deep learning model for borderline cases. It probes the model's ability to accurately classify system call sequences as malicious or benign while balancing detection latency and false positive rates in real-time scenarios. Use when the user wants to benchmark on Propedeutica System Call Dataset, or asks about evaluating this task. Reports accuracy.

Detects propaganda techniques in news articles through a two-stage pipeline: identifying text spans containing propaganda (Span Identification) and classifying the specific rhetorical technique used within those spans (Technique Classification). Use when the user wants to benchmark on SemEval-2020 Task 11 Propaganda Detection, or asks about evaluating this task. Reports F1.

Evaluates a model's ability to translate between natural language mathematics and Lean 3 formal statements (autoformalization and informalization). It measures syntactic validity, semantic correctness, and lexical similarity to assess reasoning over undergraduate-level theory. Use when the user wants to benchmark on ProofNet, or asks about evaluating this task. Reports Accuracy.

This evaluation probes a model's ability to detect prompt injection attacks in realistic deployment settings. It specifically tests whether a detector can distinguish between benign conversational or application-structured inputs and maliciously crafted injections while maintaining a very low false positive rate to avoid costly false alarms. Use when the user wants to benchmark on PromptShield Evaluation Set, or asks about evaluating this task. Reports TPR@0.1%FPR.

Evaluates Chinese large language models on five biomedical NLP tasks: information extraction, text classification, natural language inference, dialogue understanding, and content generation. It probes instruction-following, few-shot in-context learning, and parameter-efficient fine-tuning capabilities in a medical domain context. Use when the user wants to benchmark on PromptCBLUE, or asks about evaluating this task. Reports Instance-level strict micro-F1.

This benchmark evaluates the vulnerability of large language models to prompt injection attacks when generating scientific paper reviews. It probes whether hidden or biased instructions embedded in parsed PDFs can systematically skew the model's review scores and recommendations. Use when the user wants to benchmark on ICLR 2024 Review Dataset, or asks about evaluating this task. Reports Rating.

This benchmark evaluates an LLM's or monitoring system's ability to distinguish between safe user inputs and malicious prompt injection attacks. It measures both false positive rates on legitimate interactions and false negative rates on adversarial prompts to assess overall security robustness. Use when the user wants to benchmark on Gandalf, Tensor-Trust, SPML-Dataset, or asks about evaluating this task. Reports Error Rate (ER).

Evaluates 3D volumetric medical image segmentation capability on prostate MRI scans. It probes the model's ability to accurately delineate organ boundaries under clinical variability and class imbalance using end-to-end fully convolutional networks. Use when the user wants to benchmark on PROMISE2012, or asks about evaluating this task. Reports Dice coefficient.

Evaluates the fine-grained judgment capability of vision-language models by scoring generated text outputs against instance-specific rubrics and reference answers. It measures alignment with human preferences and state-of-the-art VLM judges across instruction following, VQA, and captioning tasks. Use when the user wants to benchmark on LLaVA-Bench, VisIT-Bench, Perception-Bench, OKVQA, VQAv2, TextVQA, COCO-Captions, NoCaps, or asks about evaluating this task. Reports Pearson correlation.

This evaluation protocol assesses the capability of language models to act as automated judges for other language models. It probes two distinct paradigms: direct assessment, where a model scores a single response against a reference or rubric, and pairwise ranking, where a model selects the preferred response between two candidates. The benchmarks cover instruction-following, alignment, and fine-grained custom criteria. Use when the user wants to benchmark on Vicuna Bench, MT Bench, FLASK, F...

Evaluates how well discovered motif sets in time series approximate ground truth motif sets. It penalizes false positives, false negatives, and redundant motifs without requiring uniform motif lengths or a fixed number of motif sets. Use when the user has predictions and gold and needs to compute PROM.

Evaluates a progressive feature transmission protocol for split inference at the wireless edge, measuring how efficiently features are transmitted to meet target inference accuracy or uncertainty thresholds under varying channel conditions. Use when the user wants to benchmark on GM dataset, MNIST, or asks about evaluating this task. Reports average communication latency.

Evaluates vision models on prosthesis-specific video understanding, including instance segmentation of amputees and prosthetic limbs, 2D human pose estimation with focus on lower-body keypoints, and automated gait pattern classification from pose sequences. Use when the user wants to benchmark on ProGait, or asks about evaluating this task. Reports mIoU, AP@[0.5,0.95].

This protocol evaluates the causal impact of specific profile image features (smile, body-shot, and gender) on lender selection preferences in a simulated micro-lending marketplace. It uses a conjoint-style choice experiment with GAN-generated images to isolate how visual cues influence funding decisions independent of borrower creditworthiness. Use when the user wants to benchmark on Custom GAN-generated profile images, or asks about evaluating this task. Reports Average Treatment Effect (ATE).

Evaluates reinforcement learning agents on their ability to learn efficiently and generalize to unseen, procedurally generated environments. It measures how well algorithms adapt to novel level distributions under strict computational and timestep constraints. Use when the user wants to benchmark on Procgen Benchmark, or asks about evaluating this task. Reports mean normalized return.

This evaluation probes the training efficiency and multi-GPU scaling behavior of deep learning frameworks. It measures how quickly models process mini-batches and how effectively data parallelization affects model convergence across various network architectures and hardware configurations. Use when the user has predictions and gold and needs to compute processing_time.

Evaluates multimodal foundation models on open-ended, expert-level queries across 10 professional domains, probing visual perception, domain knowledge, and long-context reasoning in single-round, multi-lingual, and multi-turn settings. Use when the user wants to benchmark on ProBench, or asks about evaluating this task. Reports ELO rating.

Evaluates real-time probabilistic forecasting of financial and weather time series, probing a model's ability to quantify uncertainty via quantile modeling and maintain calibration over sequential submission rounds. Use when the user wants to benchmark on DAX, Wind, Temperature, or asks about evaluating this task. Reports skill score.

Evaluates the skill of a probabilistic Random Forest model in forecasting severe thunderstorms (tornadoes, large hail, damaging winds) 4–8 days in advance using ensemble meteorological data. It probes the model's calibration, discrimination, and spatial coverage compared to human-generated SPC outlooks. Use when the user wants to benchmark on SPC Severe Weather Reports & GEFSv12 Reforecast, or asks about evaluating this task. Reports Brier Skill Score (BSS).

Assesses an LLM agent's ability to understand and follow privacy norms while performing real-world tasks. It measures both helpfulness and the rate at which sensitive information is incorrectly exposed. Use when the user wants to benchmark on PrivacyLens, or asks about evaluating this task. Reports privacy leakage rate.

Evaluates the trade-offs between privacy preservation, model utility, and computational/energy costs in hybrid privacy-preserving vision systems. It probes how combining federated learning with differential privacy or secure multi-party computation affects convergence, classification accuracy, and resource consumption across different neural architectures. Use when the user wants to benchmark on Alzheimer MRI Classification, ISIC Skin Lesion Classification, or asks about evaluating this task....