Eklav Models: Math Reasoning Collection Eklav checkpoints for math reasoning, plus the CotGen baseline for comparison. Same base model and data, different training method. • 10 items • Updated 28 days ago • 1
Eklav Models: Passage Reranking Collection Eklav checkpoints for passage reranking (BRIGHT, NevIR), plus the CotGen baseline for comparison. Same base model and data, different training method. • 13 items • Updated 28 days ago • 1
TabRank Qwen3 8B Table Rerankers Collection Qwen3-8B listwise table rerankers (arXiv:2607.25182). 3 variants: Naive (fast), Standard SFT (reasoning), TabRank (best). Start with TabRank. • 4 items • Updated 28 days ago • 1
CRAFT: Training-Free Cascaded Retrieval for Tabular QA Collection Data generated as augmentations to NQ-Tables and OTT-QA for Table Retrieval • 3 items • Updated 28 days ago • 1
TabRank: Chain-of-Thought Distillation for Table Re-Rankers Paper • 2607.25182 • Published Jul 28 • 1
Eklav Models Collection Eklav: hint conditioned CoT distillation. See the Passage Reranking and Math Reasoning collections below for task specific checkpoints. • 0 items • Updated 28 days ago • 1
Table + Text IR Evaluation Collection An evaluation suite created for benchmarking of retrieval models on Table+Text retrieval datasets. • 8 items • Updated Feb 17 • 2
Concrete Jungle: Towards Concreteness Paved Contrastive Negative Mining for Compositional Understanding Paper • 2604.13313 • Published Apr 14 • 12
From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agents Paper • 2603.22386 • Published Mar 23 • 57