Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2207.04802

PromptEM: Prompt-tuning for Low-resource Generalized Entity Matching

Pengfei Wang·Xiaocan Zeng·Lu Chen·Fan Ye·Yuren Mao·Junhao Zhu·Yunjun Gao
View original source ↗Read source document ↗Export metadata ↓

Abstract

Entity Matching (EM), which aims to identify whether two entity records from two relational tables refer to the same real-world entity, is one of the fundamental problems in data management. Traditional EM assumes that two tables are homogeneous with the aligned schema, while it is common that entity records of different formats (e.g., relational, semi-structured, or textual types) involve in practical scenarios. It is not practical to unify their schemas due to the different formats. To support EM on format-different entity records, Generalized Entity Matching (GEM) has been proposed and gained much attention recently. To do GEM, existing methods typically perform in a supervised learning way, which relies on a large amount of high-quality labeled examples. However, the labeling process is extremely labor-intensive, and frustrates the use of GEM. Low-resource GEM, i.e., GEM that only requires a small number of labeled examples, becomes an urgent need. To this end, this paper, for the first time, focuses on the low-resource GEM and proposes a novel low-resource GEM method, termed as PromptEM. PromptEM has addressed three challenging issues (i.e., designing GEM-specific prompt-tuning, improving pseudo-labels quality, and running efficient self-training) in low-resource GEM. Extensive experimental results on eight real benchmarks demonstrate the superiority of PromptEM in terms of effectiveness and efficiency.

Explore related subjects

cs.DB

Keep this discovery

BibTeXRIS

Pengfei Wang, Xiaocan Zeng, Lu Chen, Fan Ye, Yuren Mao, Junhao Zhu, Yunjun Gao. 2022-07-11. PromptEM: Prompt-tuning for Low-resource Generalized Entity Matching. https://arxiv.org/abs/2207.04802

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

OmniTable: A Unified Wide-Table System for Petabyte-Scale LLM Data Curation and Exploration

Data curation is a critical bottleneck in industrial-grade LLM development, where petabyte-scale unstructured corpora are scattered across hundreds of physical tables, feature engineering relies on manual, table-centric pipeline orchestration, and data lineage is largely absent. We present OmniTable as an architecture blueprint for a unified wide-table layer built on Logical Unification, Physical Separation, targeting petabyte-scale LLM data curation and exploration. OmniTable makes four contributions: (1) a unified wide-table abstraction that consolidates multi-source heterogeneous data and thousands of derived features under a single logical schema via logical-physical mapping; (2) declarative feature lifecycle management that automates dependency resolution, execution planning, operator fusion, and lineage tracking, replacing manual pipeline orchestration with a "declare-and-execute" paradigm; (3) an adaptive execution engine with autonomous governance that achieves stable PB-scale feature backfill through heterogeneous compute routing (CPU/GPU), adaptive tuning, UDF-level fault tolerance, and automated storage layout optimization; and (4) hybrid-accelerated data exploration combining a global ID index, transparent OLAP offloading, and background materialized views to deliver second-level point lookups and filtered exports exceeding 20 TB/hour. In production, OmniTable manages over 35 PB of training data across web, code, PDF, and SFT domains, reducing the human-in-the-loop curation cycle from approximately 14 days to approximately 2.5 days (5.6x over the pre-OmniTable production workflow), with consistent feature versioning, auditable lineage, and minimal manual intervention.

cs.DB↗
arXiv2026

When Does Low-Bit Quantization Preserve the Decisions of Vector Search?

Low-bit quantization can achieve high recall on some vector representations and fail sharply on others, while average distortion and global rank correlation do not explain the difference. We study quantized vector search at the level of the comparisons consumed by ranking and graph-pruning algorithms. Our first result is a distribution-free decomposition: the probability that a comparison flips is bounded by the probability mass of exact margins near zero plus the tail probability of the calibrated residual. We then account for dependence between residuals that share a query or graph node, and derive covariance-aware second-moment identities and tail bounds under a joint MGF proxy. For a frozen candidate permutation, we prove a deterministic coupling theorem for Vamana neighbour selection: the approximate replay returns the exact neighbour list exactly when all candidate-level pruning actions agree on the frozen exact states. We connect these results to representation geometry through an exact Gaussian oracle, establish a strict correlation gain from a deterministic magnitude bit in an aligned bilinear model, and give a rare-contamination construction showing why marginal Gaussian diagnostics do not imply the required residual tails. When analytical assumptions are unavailable, a held-out block certificate bounds the selective failure risk of a frozen quantized rule. Across learned, classical, and synthetic embeddings, standardized exact margins predict held-out ranking and pruning flip rates substantially better than global rank correlation. The framework applies to coordinate binary codes, RaBitQ, Lucene BBQ, and product quantizers through a common decision interface.

cs.DB↗
arXiv2026

Contextual Utility of Quantization Moves in Extreme Low-Bit LLMs

Post-training quantizers select finite code changes using reconstruction proxies or local loss approximations, but the utility of a quantization move depends on the state through which it is executed. We identify two sources of this contextual dependence. First, the displacement of the move matters: evaluating the gradient at the move midpoint captures curvature accumulated along the move that a current-state linearization omits. Across frozen two-bit moves from Llama-3.2 models, midpoint evaluation predicts the direction of exact endpoint loss changes substantially more accurately than current-state gradients. Second, moves interact: exhaustive lattices of legal quantized states are well approximated by quadratic pseudo-Boolean functions, yet their small pairwise components can determine Pareto fronts and cause different evaluation functionals to prefer opposite directions. These effects explain failures of reconstruction-optimal code re-selection and additive composition. Reading each move at its own midpoint repairs the local selection step and improves downstream accuracy and held-out perplexity, while larger supports require evaluating exact endpoints from the state actually reached. Exact-endpoint beam search finds sparse changes that dominate much larger one-shot updates, and repricing the same moves after intervening changes produces widespread sign reversals. These results show that quantization utility is contextual at the granularity of a few moves: reliable construction must evaluate finite changes along their own paths and compose them from the evolving quantized state.

cs.DB↗