A new arXiv study compares three context-engineering techniques for LLM data agents on 1,514 long-horizon tasks across five benchmarks, finding that agents without these techniques cluster into investigators, premature…

#AIagents #DataInfra #LLMs #DataEngineering
https://arxiv.org/abs/2610.20959

New arXiv work argues machine unlearning research overlooks per-sample difficulty and identifies four data/model factors that drive it. Instance-level analysis may be a useful complement to aggregate unlearning benchmarks.

#AI #MachineUnlearning #DataInfra #Research
https://arxiv.org/abs/2410.03043

A new conjoint framework called CoCo captures compactness and consistency in node representations for deep graph clustering. It combines graph convolutional filters to address local message-passing limits and reduce…

#GraphClustering #GNN #DataInfra #AIResearch
https://arxiv.org/abs/2610.11506

JevNexus reframes schema matching as bounded pairwise decisions, only invoking listwise refinement when evidence is conflicting, achieving 0.930 MRR and 0.909 Hits@1 across 561 cases while cutting mean latency from 123.452s to…

#SchemaMatching #NLP #DataInfra
https://arxiv.org/abs/2610.09487

AI agents can confidently report tasks complete while database state tells a different story, exposing gaps in agent-driven data analysis workflows. This highlights the need for stronger data infrastructure to…

#AIAgents #DataInfra #LLMs #MLOps
https://huggingface.co/blog/microsoft/thinkingbox