arXiv Machine Learning
Sep 22
Per-Query Gating of LLM Rerankers for Multi-Hop Retrieval
arXiv:2609. 22880v1 Announce Type: cross Abstract: LLM rerankers add of the order of \$0.
Retrieved but not ranked: surface-form bias in structural retrieval, from mathematics to agent trajectories
Related stories
arXiv AI
1d ago
On-Device Named-Entity Recognition: A Deployability Study of Accuracy, Cost, Reliability, and Confidence
The paper evaluates nine on‑device named‑entity recognition models ranging from classical taggers to large language models, measuring not only accuracy but also latency and output validity. Using a silver‑gold benchmark derived from an LLM judge panel and a human‑validated corpus, the study shows that encoder‑based models achieve comparable accuracy to a 4 B instruct LLM while being much smaller, faster, and producing no malformed output. Confidence calibration of GLiNER is analyzed, revealing over‑confidence but improved reliability after temperature scaling and thresholding.
arXiv AI
Aug 18
Measuring Reward Hacking and Reasoning-Answer Decoupling Under Position-Confounded Optimization
arXiv:2608. 15445v1 Announce Type: new Abstract: When a reward is correct on every training example yet consistent with more than one goal, a model can acquire an unintended one, a failure known as goal misgeneralization.
arXiv Computation and Language
Sep 24
Computation Over Geometry: Meaning Identity Is Computed, Not Shipped in the Embeddings
The paper argues that meaning identity—whether two sentences convey the same idea after wording changes—is not encoded in the geometry of independently produced sentence embeddings. Experiments on frozen off‑the‑shelf encoders and language models show that identity can only be reliably computed when both sentences are processed together in a single forward pass, yielding high accuracy (0.90–0.96) on PAWS‑X, whereas independent embeddings or simple fusion methods perform near chance. Even advanced bi‑encoder fine‑tuning improves performance on PAWS but fails to generalize to other similarity tasks, underscoring that identity is a cheap computed operator rather than a property of individual sentence vectors.
arXiv Machine Learning
Aug 13
LODESTAR: Trustworthy Entropy Is Navigated, Not Merely Measured -- Reinforced Polarizer Keeps a Frozen LLM from Being Confidently Misled by the Wrong Evidence
arXiv:2608. 11922v1 Announce Type: cross Abstract: Predictive-distribution entropy makes a strong selection rule in retrieval-augmented question answering: across five QA benchmarks, keeping the candidate answer that a frozen respondent LLM produces with the lowest answer-token entropy lifts mean answer $F_1$ from 0.
arXiv Computation and Language
Sep 22
Which Part of the Context Layer Does the Work? Separating Semantic Content from Retrieval Scaffolding in Text-to-SQL Agents
arXiv:2609.22259v1 Announce Type: new Abstract: Context layers, curated documentation that an analytics agent fetches at query time, produce large accuracy gains on text-to-SQL benchmarks. A with/wit...
arXiv AI
Sep 24
Ask Which, Not How Good: Sizing Benchmarks Scored by an LLM
The study analyzes 373,019 judgments from LLM‑scored benchmarks, decomposing variance into system, item, judge, and interaction components via generalizability theory. It finds that with a single judge, generalizability converges to a ceiling determined by the system‑by‑judge variance, which is substantially lower in pairwise preference settings, allowing one judge to suffice. The research also reveals significant biases in presentation order and highlights that many published win‑rate claims fall below the measured floor of the benchmarks.
arXiv Computation and Language
Aug 27
AgentDiff: Meaning-Bearing Rewrites Trigger Deeper Divergence than Presentation Changes in LLM Agents
The paper introduces AgentDiff, a metric that quantifies how much LLM agents’ answers differ when inputs are altered by meaning‑bearing rewrites (paraphrases, synonym substitutions) versus presentation changes (reordering, formatting, distractors). Across 68 model–benchmark–scaffold combinations involving ten LLMs and over 1,500 questions, meaning‑bearing rewrites consistently produce a roughly 20‑percentage‑point higher inconsistency rate than presentation changes, a gap that persists across severity proxies and remains significant even outside the Qwen family. Trace analysis reveals that meaning‑bearing rewrites preserve the first action but reduce thought similarity from the second step onward, extending the divergence cascade—a phenomenon termed “stealth divergence.”
arXiv AI
1d ago
Rules Amortize, Pairings Don't: Linguistic Structure Determines What Latent Task Representations Can Replace In-Context Learning
arXiv:2610.00526v1 Announce Type: cross Abstract: In-context learning (ICL) can be amortized into latent objects (task vectors, function vectors, context vectors) that recover few-shot behavior at ze...
arXiv AI
Aug 25
K-Bench: measuring model performance on real scientific agent requests
arXiv:2608.21601v1 Announce Type: new Abstract: Benchmarks for scientific artificial intelligence are mostly written to be scored: multiple-choice questions, curated agent tasks with reference soluti...
arXiv AI
Aug 17
Inducing Reward-Free Judging Rubrics that Reduce Over-Crediting in Agent Evaluation
arXiv:2608. 13564v1 Announce Type: new Abstract: Evaluating language-model agents at scale increasingly relies on a second language model as an automatic judge, because the gold signal, an executable environment reward, is expensive, slow, or unavailable at deployment time.
arXiv AI
Jun 2
Evaluating Deep Research Agents on Expert Consulting Work: A Benchmark with Verifiers, Rubrics, and Cognitive Traps
arXiv:2605. 17554v2 Announce Type: replace Abstract: Frontier deep research agents (DRAs) plan a research task, synthesize across documents, and return a structured deliverable on demand.