EEGPrep: a validated Python implementation of the EEGLAB preprocessing pipeline

Objective. Automated EEG preprocessing is common in research and clinical work, but few pipelines have been tested systematically. In a recent benchmark, the default EEGLAB pipeline was the only pipeline that significantly outperformed simple high pass filtering. Its MATLAB implementation, however, complicates deployment in Python and cloud workflows. We developed EEGPrep to reproduce this pipeline in Python while supporting BIDS data. The main technical problem was numerical: small floating point differences can accumulate during recursive filtering and ICA. Approach. EEGPrep covers the default EEGLAB workflow: average rereferencing, artifact removal with the clean rawdata plugin, Picard ICA, ICLabel component classification, channel interpolation, and epoching. We compared each stage with MATLAB reference output on ARM arm64, the primary analysis, and Intel x86 64, the supplementary analysis. The test dataset contained 64 channel P300 auditory oddball recordings from 13 participants. We measured maximum absolute difference, RMS error, AMARI distance for ICA, and agreement between ICLabel decisions. Main Results. On ARM arm64, import and rereferencing matched exactly for all 12 analysed subjects. The clean rawdata and Picard ICA stages remained at numerical zero, with maximum RMS equal to 1.5 x 10^-12 microvolts, AMARI distance less than or equal to 0.000001, and mean correlation equal to 1.000. ICLabel neural network inference introduced the only measurable difference, with maximum RMS equal to 2.0 x 10^-5 microvolts. Rejection decisions nevertheless agreed for every subject, and the difference did not increase through interpolation and epoching, with end to end maximum RMS less than or equal to 2.1 x 10^-5 microvolts. The Intel x86 64 analysis matched to the same precision. Significance. EEGPrep reproduces the validated EEGLAB pipeline and can be installed from PyPI or run in Docker.

PIPE-Cypher: Automatic Enterprise Benchmark Generation for Text-to-Cypher Systems

Enterprise property graphs vary widely in schema structure, internal terminology, domain assumptions, governance constraints, and user interaction patterns. A deployment-relevant Text2Cypher benchmark therefore reflects the questions users and agents actually ask of that graph. Creating such a benchmark is difficult because schemas and values are unique, and graph structure changes over time. Each NL-query pair must also be executable, use real graph entities, preserve diversity, and remain balanced across query types and difficulty levels. We present PIPE-Cypher, a local benchmark-generation pipeline that turns a live property graph and optional seed queries from customer questions, analyst logs, or agent tool calls into balanced NL-to-Cypher benchmarks. PIPE-Cypher combines schema profiling, reverse-query grounding, constrained generation, deterministic Cypher governance, execution validation, redaction, diversity controls, and a calibrated local LLM judge. Using local Qwen3.5-9B generation and judging, PIPE-Cypher exports 3,000 accepted FinBench/SNB examples, completes three audited ablation suites, calibrates judge behavior with human labels, and evaluates 11 local downstream models. The resulting benchmark is deliberately discriminative: zero-shot transfer is weak, while a few-shot control shows that schema-specific example banks can help compatible model families. Together, PIPE-Cypher makes Text2Cypher benchmarking a repeatable process that evolves with the graph, its users, and its target workloads.

Show, Don't TELL: Explainable AI-Generated Text Detection

Research on AI-generated text detection has presented a number of approaches to discern human from AI prose, some of which achieving high in-distribution performance. However, real-world applicability has stalled because their outputs are misaligned with the needs of users, such as professors, who are presented with a numeric score that has no attached explanation. We tackle this issue with a novel architecture, TELL, that bakes explainability from the ground-up. While our system still offers a numerical score like other detectors for comparability, TELL takes a fundamentally different approach where we aim to show the user the "tells" by which the model believes a text is AI or human-written, to empower the user to decide who wrote a text using their own judgment and understanding of the context of the writing and its alleged author. We train TELL on a custom SFT dataset of domain-specific authorship annotations, and further refine the system using GRPO with curriculum learning to improve performance. We achieve competitive performance with state-of-the-art detectors (AUROC 0.927) while natively providing annotations that explain the basis for the detector's decision. We further evaluate the quality of our explanations using a dataset of human annotations and report a high (mean 72.3%) win-rate on annotation concreteness, falsifiability, coherence, plausibility and grounding, allowing users to critically think and decide for themselves. Our work thus reframes the problem of AI-generated text detection in a human-centric perspective and paves the way for a new family of detectors that focus on native explainability.

KV Cache Quantization for Self-Forcing Video Generation: A 33-Method Empirical Study

Self-forcing video generation extends a short-horizon video model to longer rollouts by repeatedly feeding generated content back in as context. This scaling path immediately exposes a systems bottleneck: the key-value (KV) cache grows with rollout length, so longer videos require not only better generation quality but also substantially better memory behavior. We present a comprehensive empirical study of KV-cache compression for self-forcing video generation on a Wan2.1-based Self-Forcing stack. Our study covers 33 quantization and cache-policy variants, 610 prompt-level observations, and 63 benchmark-level summaries across two evaluation settings: MovieGen for single-shot 10-second generation and StoryEval for longer narrative-style stability. We jointly evaluate peak VRAM, runtime, realized compression ratio, VBench imaging quality, BF16-referenced fidelity (SSIM, LPIPS, PSNR), and terminal drift. Three findings are robust. First, the strongest practical operating region is a FlowCache-inspired soft-prune INT4 adaptation, which reaches 5.42-5.49x compression while reducing peak VRAM from 19.28 GB to about 11.7 GB with only modest runtime overhead. Second, the highest-fidelity compressed methods, especially PRQ_INT4 and QUAROT_KV_INT4, are not the best deployment choices because they preserve quality at severe runtime or memory cost. Third, nominal compression alone is not sufficient: several methods shrink KV storage but still exceed BF16 peak VRAM because the current integration reconstructs or retains large BF16 buffers during attention and refresh stages. The result is a benchmark harness, analysis workflow, and empirical map of which KV-cache ideas are practical today and which are promising research directions for better memory integration. Code, data products, and the presentation dashboard are available at https://github.com/suraj-ranganath/kv-quant-longhorizon/.

Where Bits Matter in World Model Planning: A Paired Mixed-Bit Study for Efficient Spatial Reasoning

Efficient spatial reasoning requires world models that remain reliable under tight precision budgets. We study whether low-bit planning behavior is determined mostly by total bitwidth or by where bits are allocated across modules. Using DINO-WM on the Wall planning task, we run a paired-goal mixed-bit evaluation across uniform, mixed, asymmetric, and layerwise variants under two planner budgets. We observe a consistent three-regime pattern: 8-bit and 6-bit settings remain close to FP16, 3-bit settings collapse, and 4-bit settings are allocation-sensitive. In that transition region, preserving encoder precision improves planning relative to uniform quantization, and near-size asymmetric variants show the same encoder-side direction. In a later strict 22-cell replication with smaller per-cell episode count, the mixed-versus-uniform INT4 sign becomes budget-conditioned, which further highlights the sensitivity of this transition regime. These findings motivate module-aware, budget-aware quantization policies as a broader research direction for efficient spatial reasoning. Code and run artifacts are available at https://github.com/suraj-ranganath/DINO-MBQuant.

StealthRL: Reinforcement Learning Paraphrase Attacks for Multi-Detector Evasion of AI-Text Detectors

AI-text detectors face a critical robustness challenge: adversarial paraphrasing attacks that preserve semantics while evading detection. We introduce StealthRL, a reinforcement learning framework that stress-tests detector robustness under realistic adversarial conditions. StealthRL trains a paraphrase policy against a multi-detector ensemble using Group Relative Policy Optimization (GRPO) with LoRA adapters on Qwen3-4B, optimizing a composite reward that balances detector evasion with semantic preservation. We evaluate six attack settings (M0-M5) against three detector families (RoBERTa, FastDetectGPT, and Binoculars) at the security-relevant 1% false positive rate operating point. StealthRL achieves near-zero detection (0.001 mean TPR@1%FPR), reduces mean AUROC from 0.74 to 0.27, and attains a 99.9% attack success rate. Critically, attacks transfer to a held-out detector family not seen during training, revealing shared architectural vulnerabilities rather than detector-specific brittleness. We additionally conduct LLM-based quality evaluation via Likert scoring, analyze detector score distributions to explain why evasion succeeds, and provide per-detector AUROC with bootstrap confidence intervals. Our results expose significant robustness gaps in current AI-text detection and establish StealthRL as a principled adversarial evaluation protocol. Code and evaluation pipeline are publicly available at https://github.com/suraj-ranganath/StealthRL.