TerraBench is a new benchmark with 403 tasks across Earth-science domains that evaluates LLM agents on coordinating heterogeneous data using executable ReAct-style workflows and process-level metrics.
A coefficient of agreement for nominal scales.Educational and Psychological Measurement, 20(1):37–46
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
years
2026 6verdicts
UNVERDICTED 6representative citing papers
VISTAQA is a new benchmark for joint visual question answering correctness and pixel-level grounding, evaluated with the GROVE metric that uses per-sample geometric mean to require both dimensions to succeed.
MemLens benchmark shows long-context LVLMs lose accuracy with length while memory agents lose visual fidelity, with multi-session reasoning below 30% for most systems and neither approach solving the task alone.
SPLIT benchmark finds Gemini-2.5-Flash and LLaMA-3.3-70B degrade in Ukrainian while DeepSeek-V3 stays stable, with weak human-AI agreement on cultural grounding.
LFD discovers predictive text features via LLM contrastive proposals, cross-LLM Cohen's kappa screening, and residual held-out gain selection, matching baseline accuracy while achieving higher human agreement and lower label leakage on ten tasks.
Large-scale LLM analysis of 16k CTI reports over 20 years shows a fragmented vendor ecosystem with low overlap and reporting biases.
citing papers explorer
-
TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?
TerraBench is a new benchmark with 403 tasks across Earth-science domains that evaluates LLM agents on coordinating heterogeneous data using executable ReAct-style workflows and process-level metrics.
-
VISTAQA: Benchmarking Joint Visual Question Answering and Pixel-Level Evidence
VISTAQA is a new benchmark for joint visual question answering correctness and pixel-level grounding, evaluated with the GROVE metric that uses per-sample geometric mean to require both dimensions to succeed.
-
MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models
MemLens benchmark shows long-context LVLMs lose accuracy with length while memory agents lose visual fidelity, with multi-session reasoning below 30% for most systems and neither approach solving the task alone.
-
SPLIT: Cross-Lingual Empathy and Cultural Grounding in English and Ukrainian LLM Responses
SPLIT benchmark finds Gemini-2.5-Flash and LLaMA-3.3-70B degrade in Ukrainian while DeepSeek-V3 stays stable, with weak human-AI agreement on cultural grounding.
-
Interpretable Discriminative Text Representations via Agreement and Label Disentanglement
LFD discovers predictive text features via LLM contrastive proposals, cross-LLM Cohen's kappa screening, and residual held-out gain selection, matching baseline accuracy while achieving higher human agreement and lower label leakage on ten tasks.
-
The CTI Echo Chamber: Fragmentation, Overlap, and Vendor Specificity in Twenty Years of Cyber Threat Reporting
Large-scale LLM analysis of 16k CTI reports over 20 years shows a fragmented vendor ecosystem with low overlap and reporting biases.