Evolve Quick Reviews
Public Paper Reviews
Structured preliminary reviews of recent open-access papers, generated and continuously updated by the Evolve analysis engine. Each review condenses what the paper claims, how solid the evidence is, and the questions a rigorous peer reviewer would ask — English analysis with abstracts in your language. These are independent AI analyses, not peer review; the canonical record always lives at the publisher.
Learning Where to Look: A Shared Relative-Alignment Module for Time-Series Forecasting and PPG-to-Vital-Sign Reconstruction
Ragamayi Puli, Shunya Nagashima
ROOSTER learns a comb bias replacing identity-skip and seasonal-lag conditioning; strong results, but headline wins rest on single-seed, cross-paper evidence.
Passing: An Endless Journey through Reconstructed Spacetime with AI-Generated Sound
Akira Takahashi, Chihiro Nagashima, Zhi Zhong, Shusuke Takahashi et al.
Artwork paper: slit-scan-style spacetime resampling plus real-time V2A; compelling integrated system, but no evaluation of the installation or agency claims.
NV-Reason-CT: 3D Visual Language Model for CT Analysis
Andriy Myronenko, Dong Yang, Yucheng Tang, Baris Turkbey et al.
Native-3D CT VLM (CT-RATE F1 0.614) with rigorous baselines; promising, but Merlin test-set prompt tuning and a thin human study temper the claims.
Uncheatable Eval: Dynamic Compression-Based Evaluation of Language Models
Kaifeng Tan, Yudong Li, Linlin Shen
Dynamic compression benchmark on fresh text for 80 base LLMs; scaling, long-context and MMLU findings are solid, though correlations lack size controls.
PhyMo: A Physical-Field Modality for Multimodal AI4Physics
Henan Sun, Haitao Hu, Jin Liu, Jianfeng Zhang et al.
Claims PDE-structured physical fields as a new multimodal modality for AI4Physics; promising multi-dataset gains but thin margins, trivial theory, no code.
Pistis Technical Report
Heyun Chen, Xiaohan Lan, Jiaxi Li, Zhilin Lu et al.
Interleaved distillation+RL post-training gives stable training and search gains over Qwen bases, but ablation deltas are small and no artifacts are released.
DeltaS: Reading the Gated Linear Attention State for KV Cache Eviction in Streaming Video
Taeyoun Kwon, Seungjin Kim, Hyeonyu Kim, Moon Hwan Kim
Proposes gated-delta state drift as a free KV-eviction signal for streaming video; controlled gains are solid, but single backbone and no significance tests.
Not What You Meant: Can LLMs Follow a Specified Negation Semantics?
Qiming Bao, Agnieszka Mensfelt, Michael J. Witbrock, Kostas Stathis
严格的求解器认证基准测试表明，Frontier 下的 LLM 往往无法遵循指定的否定语义；合成范围限制了普遍性。
WhatWorkedBench: Benchmarking Experimental Understanding in AI Agents
Jingjie Ning, Xueqi Li, Yibo Kong, Dongting Li
Benchmarks agents' predictions of component effects vs exhaustive execution; rigorous, reproducible design, but evidence limited to one provider's two models.
Behaviora - A Conceptual Architecture for External and Internal Behavior of Robots and Agents
Gote Nyman
Proposes behavior as shareable, addressable computational entities (episodes + IoB addresses); honest but untested conceptual work with hand-crafted arm demos only.
EBRL: Asynchronous Embodied RL by Multi-Grained Resource Management
Liang Mi, Weijun Wang, Bowen Gao, Tianze Yu et al.
EBRL claims 1.30-3.47x embodied-RL throughput via async pipelining and elastic CPU/GPU allocation; broad, consistent eval, but single-run stats limit rigor.
Robustness of Diffusion Models under Distribution Shift
Wei Luo, Neil K. Chada, Shijie Zhang, Lu Yu
Proves minimax-optimal robust score/generation theory for OU diffusion under Wasserstein shift; rigorous, but confined to OU dynamics and the small-t0 regime.
ProCredit: From Outcome Rewards to Progress Credit in Agentic Reinforcement Learning
Ming Ma, Yi Zhu, Yiran Zhong, Feida Zhu et al.
Turns verifiable acceptance checks into per-turn progress rewards for agentic RL; consistent 3-scale AppWorld gains, clean ablations, but verifier-bound scope.
M3D-Net: Hierarchical Coordination of Spatial Context, Feature Reuse, and Differential Attention for Mammography Classification
Zheng Yu, Xinhang Li, Jiabao Gao, Boyang Wang et al.
Higher mammography val accuracy (97.78%) from a three-mechanism encoder, but single-seed runs, tiny data, and no ablations leave the claims unverified.
ThaiTrees: Thai Syntactic Dependency Trees Across Domains
Attapol T. Rutherford, Papatchol Thientong
一项宝贵的 342M-token 解析泰语语料库，涵盖了四个领域；虽然看起来合理且描述坦诚，但缺乏输出质量评估（QA），并且未审视关键词分析（keyness）中的解析伪影风险。
DCRL: Decoupling and Coupling Reinforcement Learning via Policy-Reward Manifold Alignment
Henan Sun, Zehua Li, Haitao Hu, Qifan Zhang et al.
DCRL co-evolves policy and reward via shared weights + evolving rubrics; a 4B model beats 32B, but theory is thin and key ablations missing.
Hidden not Deleted: How Networks Suppress Entangled Features
Akash Samanta, Manish Pratap Singh, Debasis Chaudhuri
Gradient concept erasure hides rather than deletes entangled features, bifurcating into mirror/shadow solutions; causally validated but toy-scale only.
TNLearn: An Open Source Python Package for Task-based Neurons
Meng Wang, Tieyun Li, Juntong Fan, Hanyu Pei et al.
TNLearn: an open-source package for auto-designed task-specific neurons; valuable tool, but evidence rests on single runs with possible baseline under-tuning.
FDE-Bench: Evaluating LLM Agents for Deployment Environment Configuration
Weihang Ding, Junfei Zhan, Yueting Li, Qirong Guo
Rigorous replay-graded benchmark for LLM deployment agents with strong anti-gaming gates; single-episode cells and n=1 human baseline limit statistical power.
BaseCamp --- An Agentic AI Framework for Automating DNA Sequencing Data Pipelines
Eranga Bandara, Xueping Liang, Asanga Gunaratna, Tharaka Hewa et al.
BaseCamp proposes six LLM agents for the decision layer of sequencing pipelines; sensible design, but every reported result is an unmeasured placeholder.
Does Step Law Transfer to Small-Scale Language Models? An Empirical Recalibration Below 59M Parameters
Egor Romanyukov, Timofey Novikov, Timur Shokarov, Elizaveta Zorkina et al.
Step Law LR ~4x too high below 3M params; recalibrated power laws fit; promising but single-seed, confounded by protocol changes from the original
Efficient Linear Bandits via Cluster-Aware Sketching
Hantao Yang, Hong Xie, Defu Lian
Clever lossless per-cluster sketching yields fixed-size-sketch linear bandits with converging regret; but regret scales with √N and DBSLinUCB is untested.
When Context Misleads: In-context Learning with Jurisdiction in Large Language Models
Pei-lin Li, Qingle Liu, Junyang Feng, Siyu Li et al.
Benchmark + J-ICL post-training claim ICL and fake-context resistance improve together; well-audited but single-seed, no CIs, thin Qwen3-4B gains.
Who Is Behind the Harness? Fingerprinting LLMs through Agentic Behavior
Chuyi Wang, Xiaohui Xie, Tongze Wang, Fangchen Luo et al.
主动式黑盒大模型指纹识别，通过编码智能体探针实现；在 36 个模型上表现出色，并进行了消融实验，但仅限于闭集（closed-set）场景，且规避测试（evasion untested）尚未验证。