search engineers

Abstract: Hierarchical text embedding models offer variable-length representations, while quantization reduces numerical precision; both aim to improve computational efficiency. This work is an empirical benchmark study of how these two orthogonal, already-established techniques interact. We evaluate 38 tasks from the Massive Text Embedding Benchmark (MTEB) using four transformer-based embedding models, covering 1D and 2D hierarchical representations combined with 32-bit, 8-bit, and binary quantization, for 168 configurations per task. We report each efficiency dimension separately: theoretical per-vector storage and end-to-end evaluation time. Under a Friedman test with Nemenyi post-hoc analysis, int8 quantization is statistically indistinguishable from float32 for pair classification and re-ranking and costs at most 1.5 percentage points overall, whereas binary quantization is significantly worse across every task category. Halving the embedding dimension once is not statistically detectable above $m=16$; two successive halvings always are. Combining an 8x dimensionality reduction with int8 yields a 32x reduction in per-vector storage at 94.4% of baseline effectiveness. Because evaluation times were measured once, we treat them as indicative: we find no evidence that quantization speeds up evaluation, whereas sampling from intermediate layers shortens it substantially. Retrieval degrades far faster than every other category, so we report findings per task category.

Abstract: Algorithm-driven services, including search, recommendation, and advertising, rely on multi-stage model pipelines. While research prioritizes model optimization, post-deployment monitoring focuses on operational reliability. Yet operationally healthy systems may still produce poor-quality results, exposing a gap in existing evaluation and monitoring practices. However, a systematic problem formulation and benchmark for this task remain lacking. To bridge this gap, we formulate System Quality Diagnosis (SQD) for algorithm-driven services, a new task that extends quality evaluation beyond identifying poor outcomes to attributing their causes across system components. Focusing on industrial search, we propose WISE, an agentic framework for end-to-end quality diagnosis that discovers user-facing issues and traces their origins to underlying algorithmic components. To improve diagnostic reliability, we introduce ReTrace for constructing evidence-consistent trajectory supervision and Adaptive Harness-Policy Optimization (AHPO) for failure-guided optimization of the agent harness and model policy. To enable systematic evaluation, we construct WISE-BENCH, a real-world multimodal search benchmark that combines user-facing evidence, internal pipeline traces, and expert diagnosis annotations. Extensive experiments show consistent gains across model scales, with WISE achieving 75.20% end-to-end accuracy on Qwen3.5-35B-A3B. In production, WISE supports quality auditing, online quality assurance, and model optimization, reducing diagnosis costs and improving downstream relevance models.

Abstract: Chunking is a fundamental preprocessing step in retrieval-augmented generation (RAG), yet most existing chunkers rely primarily on heuristics. We introduce the Question--Evidence-Supervised Chunker (QESChunker), to our knowledge the first method for constructing static chunk indices from question--evidence supervision. QESChunker constructs document-specific chunk indices by optimizing over human-annotated or LLM-generated question--evidence pairs, without access to held-out test queries. We formulate non-overlapping and overlapping index construction through a unified question--evidence utility-maximization objective, turning chunking into a supervised combinatorial optimization problem. Under the non-overlap constraint, the objective decomposes over chunks and can be optimized exactly in polynomial time by dynamic programming. When overlap is allowed, the objective becomes monotone submodular; exact optimization is NP-hard, but our greedy achieves a $(1-1/\sqrt{e})$-approximation. Across two self-constructed and three public datasets, both variants outperform existing chunking methods in complete-evidence recall in nearly all evaluated settings, under the same overlap regime and a fixed hybrid retriever. These results recast chunking as supervised, budget-aware index construction for evidence retrieval.

Abstract: We report a systematic dissociation in modern text embedders: sentiment is almost perfectly decodable from their embeddings, yet almost entirely absent from their metric. Across eleven frozen encoders spanning 2021 to 2026, including 2026-generation models and the 4B-parameter LLM embedder Qwen3-Embedding-4B, a linear probe recovers sentence polarity at up to 91.3% accuracy while cosine similarity separates polarity by a margin of only 0.003-0.064. Qwen3-Embedding-4B, the strongest probe in the panel, reaches 91.3% probe accuracy but only 64.3 retrieval MAP. Because retrieval, clustering, deduplication, and retrieval-augmented generation operate on the metric rather than on a trained probe, these systems are effectively blind to information their encoder demonstrably contains. A task-conditioning control suggests this reflects the pretraining objective rather than a prompting artifact: conditioning Jina-v5 on its classification task rather than its retrieval task recovers only about half of the MAP gain that injection achieves on the same weights (63.2 to 73.8, versus 85.1), and leaves the alignment gap essentially unchanged. Sentiment-Injected Sentence Embeddings (SISE) train a small (0.13M-0.69M-parameter) supervised-contrastive head on a frozen encoder in seconds on a CPU, using the labelled training set. The head improves every encoder released since 2022 on full-data probe accuracy (up to +5.8 points), retrieval MAP (up to +26.0) and polarity margin (7-61x). This is a specialization, not a free gain: the compact 128-d head loses 38-41 Spearman points on STS-B, while a fused variant that concatenates it with the base embedding loses 2.4-12.0. With the head, a 0.6B encoder at 128 dimensions matches a 4B encoder given the same head (91.8 vs. 90.3 MAP). We release a CPU-reproducible pipeline and a reusable diagnostic, the Attribute Alignment Gap.

Abstract: Meaning identity (whether two sentences say the same thing after wording changes) is treated in retrieval and RAG as a geometric fact about independently encoded sentence vectors. We show that, for frozen off-the-shelf encoders and language models, it is not: identity is computed when both sentences share one forward pass, and is not a property of the embedding geometry those systems ship. On overlap-matched PAWS-X, purpose-built encoders (BGE, E5, GTE, MiniLM, E5-Mistral-7B) reach English confirm AUC only 0.55-0.65 (dense peak 0.70). Independently encoded last-token states of Llama 3, Mistral, and Qwen do no better; late fusion of the two vectors stays near chance. The same probe on a joint forward pass reaches 0.90-0.96 from 1.5B to 32B, collapses under partner shuffle, is mid-depth, saturates near 0.94 by 3B, and appears more weakly in GPT-2 XL (0.76). The gap holds beyond Llama-style models on other causal LMs, bidirectional encoders (DeBERTa, RoBERTa), and encoder-decoders (Flan-T5, T5, BART). Fixed or linear readers over frozen independent encodings never unlock identity; nonlinear pair readers recover part of it only on the full 49k-pair PAWS train split (0.68-0.87). Off-the-shelf rerankers split: BGE-reranker-large reaches 0.94, while MS-MARCO and Jina stay at 0.55-0.64. Independently trained families compute the same relation and a 1.5B joint reader can distill it from unlabelled teacher scores, while no linear function of the teachers own independent vectors can. Bi-encoders can be fine-tuned to fit PAWS (0.87-0.93), but transfer and STS-B suffer. Cosine compares wording neighbourhoods; identity is a cheap computed operator, not a property of either sentence vector.