arXiv AI
Jun 2
Query Circuits: Explaining How Language Models Answer User Prompts
arXiv:2509. 24808v2 Announce Type: replace Abstract: Explaining why a language model produces a particular output requires local, input-level explanations.
Beam Search, Self-Consistency, and the Limits of Inference-Time Scaling for Grammar-Constrained Text-to-SQL in Small Language Models
The paper investigates how increasing inference-time computation—via wider beam search or sample‑plus‑vote—affects performance on grammar‑constrained text‑to‑SQL tasks for small language models. Using the Qwen2.5‑Instruct family (0.5B–7B parameters) on the Spider benchmark, the authors find that larger models consistently outperform higher inference compute on the same model size, and that beam search yields better accuracy than sample‑plus‑vote under matched budgets. These results suggest that, unlike unconstrained settings, scaling inference compute does not compensate for smaller model size when strict grammar constraints are applied.
Related stories
arXiv Machine Learning
Aug 12
The Parser Already Knows: Lightweight Bias Correction in Constrained Decoding
arXiv:2608. 10137v1 Announce Type: cross Abstract: Grammar Constrained Decoding (GCD) forces Language Models (LMs) to produce syntactically valid outputs by masking out non-conforming tokens at each step.
arXiv Computation and Language
Sep 17
Register Bias in Complexity-Based Large Language Model Routing
The paper examines how large language model (LLM) services route queries to models of varying size based on a cheap complexity estimate. It finds that this routing is not register neutral: queries written in non‑standard English registers (e.g., African American English or second‑language English) are systematically assigned to lower‑capacity models because they appear shorter due to omitted function words. Experiments on 37,704 learner sentence pairs and a controlled corpus show that this bias leads to significantly lower accuracy across all model tiers, including the highest‑capacity cloud models, while the routing decision itself adds little marginal cost.
arXiv AI
Jun 30
Field Order Should Not Matter: Permutation-Invariant Embedding Model Fine-Tuning for Structured Metadata Retrieval
arXiv:2606. 30473v1 Announce Type: cross Abstract: We study retrieval over catalogs of structured metadata, where each record is a small schema whose fields answer different kinds of query.
arXiv AI
Jun 26
The Capability Frontier: Benchmarks Miss 82% of Model Performance
arXiv:2606. 26836v1 Announce Type: new Abstract: Existing benchmarks typically report accuracy for a single model on a single run.
arXiv AI
Aug 14
Large Language Models Can Follow Instructions, But Not Many at Once: Phase Transitions in Compositional Constraint Satisfaction
arXiv:2608. 12426v1 Announce Type: new Abstract: Large language models are increasingly deployed in settings that require simultaneous adherence to multiple explicit constraints - reasoning structure, safety boundaries, output schemas.
arXiv Computation and Language
Sep 15
Mixture-of-Experts Language Models Can Be Strong and Efficient Retrievers
arXiv:2609.13486v1 Announce Type: cross Abstract: Recent work has shown that fine-tuning decoder-only large language models (LLMs) for retrieval yields strong first-stage retrievers, with effectivene...
arXiv Computation and Language
2d ago
Making Grid Beam Search Less Greedy
arXiv:2609.39368v1 Announce Type: new Abstract: A common formalism for constraining the output of autoregressive text generation models involves lexical constraints, words or phrases which are requir...
arXiv AI
Jun 24
CAVEWOMAN: How Large Language Models Behave Under Linguistic Input and Output Compression
arXiv:2606. 24083v1 Announce Type: cross Abstract: "Talk short.
arXiv AI
Jul 13
Test-Time Scaling for Small VLMs on Multilingual Visual MCQ
arXiv:2607. 09438v1 Announce Type: cross Abstract: Test-time scaling (TTS) reliably improves reasoning in large language models, but whether it transfers to small open vision-language models remains unclear.
arXiv AI
Jul 1
Optimal Self-Consistency for Efficient Reasoning with Large Language Models
arXiv:2511. 12309v2 Announce Type: replace-cross Abstract: Self-consistency (SC) is a widely used test-time inference technique for improving performance in chain-of-thought reasoning.
arXiv AI
3d ago
HARISSA: Inference-Time Self-Checks for Efficient and Safe Local Language Model Deployment
arXiv:2609.38006v1 Announce Type: new Abstract: Running a language model locally offers advantages in privacy, latency, and cost, but local hardware fits only small models, which are less capable tha...