ALTK‑Evolve: On‑the‑Job Learning for AI Agents
•
16
AI & ML interests
Enterprise AI and ML, Foundation Models, Responsible AI
Recent Activity
View all activity
Papers
VAREX: A Benchmark for Multi-Modal Structured Extraction from Documents
Look Where It Matters: High-Resolution Crops Retrieval for Efficient VLMs
Articles
ibm-research/ScarfBench
Updated • 1.79k • 5
ibm-research/AssetOpsBench
Viewer • Updated • 467 • 600 • 20
ibm-research/VAKRA
Viewer • Updated • 1.33k • 1.16k • 40
ibm-research/data-product-benchmark
Viewer • Updated • 33.2k • 193 • 3
ibm-research/900K-Judgements
Viewer • Updated • 939k • 106 • 2
ibm-research/VAREX
Viewer • Updated • 1.78k • 1.44k • 3
ibm-research/BPO-Bench
Preview • Updated • 31 • 1
ibm-research/watsonxDocsQA
Viewer • Updated • 1.22k • 109 • 5
ibm-research/AITQARetrieval
Viewer • Updated • 3.99k • 90
ibm-research/ITBench-Lite
Updated • 670 • 5
ibm-research/flow-bench
Viewer • Updated • 546 • 16 • 2
ibm-research/FailureSensorIQ
Viewer • Updated • 8.3k • 188 • 6
ibm-research/WatsonxDocsQARetrieval
Viewer • Updated • 1.2k • 20
ibm-research/StatCanDialogueRetrieval
Viewer • Updated • 7.65k • 5
ibm-research/MultiHierttRetrieval
Viewer • Updated • 11.9k • 41 • 1
ibm-research/OTTQASmallRetrieval
Viewer • Updated • 31.1k • 37 • 1
ibm-research/NQTablesRetrieval
Viewer • Updated • 533k • 92 • 2
ibm-research/OpenWikiTablesRetrieval
Viewer • Updated • 178k • 60 • 1
ibm-research/FeTaQARetrieval
Viewer • Updated • 24.7k • 84
ibm-research/rag-hpo-bench
Preview • Updated • 2.04k • 2
ibm-research/acp_bench
Viewer • Updated • 3.72k • 1.61k • 13
ibm-research/dp-bench
Updated • 18 • 1
ibm-research/ITBench-Trajectories
Updated • 207 • 3
ibm-research/nc-bench
Viewer • Updated • 720 • 125 • 1
ibm-research/ReasonLab
Viewer • Updated • 4.35k • 444 • 1
ibm-research/Auto-BenchmarkCard
Viewer • Updated • 105 • 29 • 3
ibm-research/MermaidSeqBench
Viewer • Updated • 132 • 37 • 5
ibm-research/knowledge_consistency_of_LLMs
Preview • Updated • 35 • 3
ibm-research/ToolRM-train-data
Viewer • Updated • 459k • 129 • 7
ibm-research/BFCL-FC-robustness
Updated • 38