arXiv Machine Learning

Beyond Endpoint Scores: Time- and Capacity-Conditioned Evaluation of Continual Knowledge Updating

The paper argues that evaluating continual knowledge‑updating methods solely at a final checkpoint and a single adapter rank can be misleading. By fixing a periodic hierarchy and comparing it to cumulative replay on a 24‑month Wikidata stream, the authors show that the apparent best method changes depending on the evaluation month, replay LoRA rank, and query formulation. They recommend reporting performance trajectories and capacity sweeps, and only declaring a robust winner when the ranking remains stable across the evaluation region.

arXiv AI
2d ago

LSREP: A Longitudinal State-Replay Protocol for Evaluating Conversational Memory, with ICE v2 as an Audited Local-First Architecture

The paper introduces LSREP, a Longitudinal State‑Replay Evaluation Protocol designed to assess how conversational memory evolves over time, incorporating ordered replay, lifecycle schedules, repeated probes, evolving reference answers, and mechanism‑fidelity checks. It applies LSREP to ICE v2, a local‑first memory middleware, and reports that on three ordinary‑density datasets ICE v2 achieves near‑zero mean quality difference from vector‑RAG while using fewer fragments but slightly more prompt tokens, yet fails catastrophically on a dense dataset. In a public diagnostic, ICE v2 underperforms pure vector‑RAG on LongMemEval, revealing significant multi‑session and temporal failures and a quality‑cost trade‑off rather than superior efficiency.

By Deepesh Sonar
Hugging Face Trending Papers
Sep 3

What Matters for Aggressive Decoding-Time KV Eviction? Temporal Aggregation and Ranking Preservation

The paper investigates how temporal aggregation of token scores during decoding-time KV cache compression affects eviction decisions. It shows that using an exponential‑moving‑average (EMA) aggregation keeps ranking changes minimal for many scorer variants, while others like KeyDiff or recency significantly alter rankings and degrade performance. Building on this, the authors propose InertiaKV and its lazy variant, which achieve 1.34–1.46× faster decoding than full refresh, and also evaluate a score‑free approach that slightly improves quality while eliminating further scoring.

arXiv AI
Sep 4

What Matters for Aggressive Decoding-Time KV Eviction? Temporal Aggregation and Ranking Preservation

The paper investigates how temporal aggregation and ranking preservation affect aggressive decoding-time KV cache eviction. It shows that exponential‑moving‑average (EMA) aggregation makes many scorer modifications indistinguishable at the eviction‑set level, while certain variants (KeyDiff, key norm, recency, learned scorer) alter rankings and degrade performance. Building on this, the authors propose InertiaKV and its lazy variant, achieving 1.34–1.46× faster decoding, and also explore a score‑free decoding approach that slightly improves quality while eliminating further scoring.

By Bo Zeng, Yu Zhao, Yefeng Liu, Zhihong Lu, Xuanfan Ni, Xintong Wang
arXiv AI
Aug 26

Matched Excess-Outranker Regularization for Candidate-Set Interference in Continual Knowledge Graph Embedding

The paper introduces Matched Excess-Outranker Regularization (MEOR), a new host-level objective for continual knowledge graph embedding that addresses candidate-set interference caused by entity admission. MEOR compares newcomer pressure with matched old references, applying a one-sided penalty only when newcomers outcompete these references, thereby preserving the learner’s signal for legitimate new entities. Experiments on ENTITY-ComplEx and FBInc datasets show that MEOR improves historical current-universe mean reciprocal rank and reduces candidate-set interference, outperforming several baseline regularizers.

By Hao Ren, Junbin Gao, Jiaojiao Jiang
arXiv AI
Aug 24

UpgradeBench: A Decision-Centric Benchmark for Upgrading Fine-Tuned LLM Specialists

UpgradeBench is a decision‑centric longitudinal benchmark that evaluates how fine‑tuned language‑model specialists should be handled when new base‑model releases occur. It covers four consecutive Qwen releases, a continuation checkpoint, six tasks, two model sizes, and OLMo checkpoints with known training lineage, and examines whether retraining, adapter transfer, or other recovery strategies improve specialist performance. The benchmark reveals that upgrade gains vary by task and release interval, that direct adapter copying is sensitive to pretraining distance, and that teacher relabeling can recover specialists without new annotations. "whyItMatters":"The study provides actionable insights into the cost‑effective management of specialist models across model releases, showing how to balance retraining effort with performance gains."

By Ye Chen, Weining Zhang
arXiv AI
Aug 28

Why RAGs Hallucinate: Penalty-Aware Evaluation of Retrieval-Augmented Generation Systems with Knowledge-Gap Canaries

The paper introduces a penalty‑aware evaluation framework for Retrieval‑Augmented Generation (RAG) systems that uses asymmetric scoring, knowledge‑gap canaries, and a failure‑attribution pipeline. Applying this framework to three commercial RAG products and a baseline on SimpleQA‑Verified, the authors find that while overall accuracy is similar across systems, canary violation rates vary dramatically, showing that systems differ more in when they answer than in what they answer. The study demonstrates that penalty‑aware scoring can reorder system rankings and is robust across different penalty settings.

By Alden Do Rosario, Hussein Younes, Felipe Pires