Papers for

ai benchmarking teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Embodied-BenchForge builds better robot task benchmarks with error checking

Embodied-BenchForge: A Closed-Loop Agentic Workflow for Embodied Benchmark Construction

Abstract: Agentic systems offer a promising way to automate embodied benchmark construction, but existing approaches typically cover isolated stages or remain specialized to predefined environments and task families. More importantly, multi-step construction produces dependent intermediate artifacts that are often passed downstream without artifact-specific verification, allowing local defects to propagate into the final benchmark. We present Embodied-BenchForge, an agentic framework that transforms user-specified evaluation intents into complete embodied benchmark artifacts. It formulates construction as Closed-Loop Benchmark Synthesis, integrating forward artifact synthesis with backward verification and repair. Skill-Orchestrated Artifact Synthesis composes typed and reusable skills into executable workflows, while an artifact dependency graph records intermediate outputs and their dependencies. Requirement-Guided Verification and Repair applies artifact-specific contracts throughout construction and uses provenance to trigger local re-execution or upstream rollback when verification fails. Embodied-BenchForge constructs six benchmarks covering diverse embodied scenarios in the Offline EQA Track, together with one interactive benchmark containing 220 executable tasks in the Interactive Embodied Track. Evaluations of representative MLLMs and embodied agents show that the benchmarks distinguish model capabilities in both observation-based understanding and closed-loop execution. Quality assessment and ablations validate benchmark quality and the effectiveness of verification and repair, while repair and skill-reuse analyses demonstrate efficient localized recovery and cross-benchmark reusability.

Fri 11 SeptArtificial Intelligence
The gist
Creating complex tests for robots and AI that interact with the world is hard because mistakes in early steps can cause big problems later. The authors present Embodied-BenchForge, a system that not only builds these robot task tests but also double-checks and fixes errors along the way. It uses a clear process where tasks are made, checked, and repaired to ensure high-quality benchmarks that help better evaluate robot and AI abilities. They showed this system can create several diverse test sets and improve reliability compared to previous methods.
Open 2609.13082v1

Cybersecurity model scores vary widely with evaluation methods

Benchmark Scores Are Pipeline-Dependent: A Reliability Audit of Cybersecurity LLM Benchmarks

Abstract: Large language model (LLM) benchmarks are often treated as fixed datasets with stable scores, yet their outcomes depend on configurable evaluation pipelines. We audit eight cybersecurity benchmarks across 10 proprietary, open-weight, and cybersecurity-specialized LLMs. By modeling benchmarks as measurement pipelines, we identify 15 systematic failure modes and show that a single pipeline choice can change a model's score by more than 80 percentage points and substantially alter model rankings. At the cross-benchmark level, two semantically similar task pairs rank the same models differently because of incompatible evaluation conventions. Under an evaluation harness that standardizes pipeline choices while preserving task semantics, nine of 10 models shift by at least three ranks on at least one benchmark. These results show that cybersecurity LLM benchmark scores are pipeline-dependent and motivate pipeline-aware auditing as a core requirement for reliable model evaluation.

Tue 8 SeptCryptography and SecurityArtificial IntelligenceComputation and Language
The gist
When testing cybersecurity language models, the way tests are set up can drastically change the results. The authors found that changing evaluation choices can shift a model’s score by over 80 points and reorder model rankings. Even tasks that seem similar can rank models differently if evaluated differently. Their work shows the importance of carefully standardizing how these tests are done to get reliable comparisons.
Open 2609.08765v1