Search: “models”

Sources: Google plans to release Gemini 3.8 Flash as soon as Wednesday; Gemini 4 has done well on pre-training evals but still needs to complete post-training (Erin Woo/Wall Street Journal)

Erin Woo / Wall Street Journal: Sources: Google plans to release Gemini 3.8 Flash as soon as Wednesday; Gemini 4 has done well on pre-training evals but still needs to complete post-training — Interna

model-releasestechmeme

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Stress-Testing Efficient Responsible-AI Evaluation: When Compute Savings Change Benchmark Conclusions

arXiv:2608.31108v1 Announce Type: new Abstract: Efficient evaluation changes the protocol used to support claims about model behavior, yet it is rarely tested whether those claims remain stable after

Sycophantic Agreement Transfers with Neutral Data via Contrastive Preference Optimization

arXiv:2608.31079v1 Announce Type: new Abstract: Sycophantic agreement refers to a behavior in which language models excessively affirm the user, often at the cost of factual accuracy. Although sycopha

safetyarxiv-cs-lg

TAKE 85: Testing Audiovisual filmmaKer's intEnt across 85 Hours of Film

arXiv:2608.30068v1 Announce Type: new Abstract: Films communicate through deliberate creative choices, including lighting, color, composition, editing, dialogue, music, and sound. Humans naturally int

Train past the frontier. @tryheidi's ambient scribe drafts a clinical note the clinician has to trust enough to sign. They fine-tuned open m…

model-releasesfireworks-ai--x

Train past the frontier. @tryheidi's ambient scribe drafts a clinical note the clinician has to trust enough to sign. They fine-tuned open models on Fireworks whose notes their clinicians preferred ov

Training and Agentic Inference Strategies for LLM-based Manim Animation Generation

arXiv:2604.18364v2 Announce Type: replace Abstract: Generating programmatic animation using libraries such as Manim presents unique challenges for Large Language Models (LLMs), requiring spatial reaso

Automatic Pronunciation Error Detection and Correction of the Holy Quran's Learners Using Deep Learning

arXiv:2509.00094v2 Announce Type: replace-cross Abstract: Assessing spoken language is challenging, and quantifying pronunciation metrics for machine learning models is even harder. However, for the H

CommerceVibe: Learning to Design E-Commerce Creatives as Executable Visual Code via Dual-Feedback Reinforcement Learning

arXiv:2608.27893v1 Announce Type: new Abstract: High-quality e-commerce creatives are essential for presenting products and conveying marketing messages. Recent diffusion models enable scalable creati

Does Finetuning with Scientific Data Increase Hallucinations? A Multi-domain Factuality Evaluation of LLMs

arXiv:2606.21359v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used to communicate and explain scientific concepts, yet their tendency to hallucinate poses significa

FinExam-10K: When Retrieval Helps Financial Reasoning?

arXiv:2608.28155v1 Announce Type: new Abstract: Professional financial examinations require models to combine domain knowledge, calculation, and judgment, yet no benchmark covers the full CFA and FRM

Generalized Gibbs Ensemble Weighting for Forecast Combination

arXiv:2608.28116v1 Announce Type: new Abstract: Forecast combination is a reliable way to improve predictive performance when several forecasting models are available. Simple aggregation rules such as

INSPIRE: An Internalize-Then-Improve Approach for Example-Driven Mathematical Reasoning

arXiv:2608.27501v1 Announce Type: new Abstract: Mathematical reasoning has seen rapid progress in large language models (LLMs), yet existing methods optimize predominantly for final-answer correctness

safetyarxiv-cs-cl

Research

Layered LLM Defenses as an Ensemble: Access Tiers, Inference Cost, and the Measured Failure Correlation Between Defense Layers

arXiv:2608.28327v1 Announce Type: cross Abstract: Practitioners defend large language models (LLMs) by stacking defenses, assuming the layers compound. A stack is an ensemble, and ensembles compound o

researcharxiv-cs-cl

Prequential posteriors

arXiv:2511.17721v2 Announce Type: replace-cross Abstract: Data assimilation is a fundamental task in updating forecasting models upon observing new data, with applications ranging from weather predict

Qwen3.8-Flash-Next in llama.cpp from CPU-only to 96GB VRAM: 8.5 to 109 tok/s, max context and parameters test. My findings on RTX 6000 PRO.

Hey guys, I tested Qwen3.8 Flash with llama.cpp from CPU-only to the full 96GB of my RTX PRO 6000. Short version: CPU-only reached 8.34 tok/s at a 2K prompt Full 96GB reached 109.07 tok/s At 245K cont

Residual-Guided Randomized Neural Networks

arXiv:2608.28267v1 Announce Type: new Abstract: Randomized neural networks enable fast and analytically tractable training by fixing the input to hidden layer parameters at random and learning the out

Riverbank Erosion Analysis in Bangladesh Using Spatiotemporal Segmentation

arXiv:2510.17198v2 Announce Type: replace Abstract: Riverbank erosion is a serious environmental problem in Bangladesh, causing land loss, damage to infrastructure, and displacement of local communiti

Robust Chance-Constrained Optimization using a Continuous Parameter Space Wasserstein-2 Ambiguity Set of Gaussian Mixtures

arXiv:2607.17018v2 Announce Type: replace-cross Abstract: We study distributionally robust linear chance-constrained problems in which uncertainty is modeled by a Gaussian mixture model (GMM). Finite-

SciReC: Diagnostic Evaluation of Multimodal, Multi-Turn Relational Reasoning with Adaptive Interaction

arXiv:2608.27461v1 Announce Type: new Abstract: Relational reasoning requires the process of perceptual understanding, comparing, and integrating the underlying relationships between concepts. This ab

Hardware

SOMTab: Set-Order Mamba for Efficient Tabular In-Context Learning

arXiv:2608.27882v1 Announce Type: new Abstract: Tabular foundation models based on in-context learning have recently emerged as strong alternatives to task-specific model fitting. However, the current

hardwarearxiv-cs-lg

What Can Low Resource Languages Learn From Each Other?

arXiv:2608.27753v1 Announce Type: new Abstract: Despite the rapid advancement of Vision-Language Models (VLMs), their linguistic reach remains largely confined to high-resource languages, leaving the

I collected every single LLM coding benchmark, and computed their Intelligence Density

The intelligence in my context is an aggregate index, I called the Agentic Coding Index, across most relevant agentic coding benchmarks: SWE-bench Pro, DeepSWE v1.1, Terminal-Bench (v4, v3, v2.1), Cod

30 Aug 2026

Unpopular opinion Qwen 3.8 is hard to understand

I find both Qwen 3.8 27b and Qwen 3.8 Flash Next difficult to read. Here's some examples of what I mean: **Model-visible tool set per turn** (assembled by the host at provider-request time): persona t

30 Aug 2026

ADeptS-Bench: Measuring the Trustworthiness of Computer Use Agents Across Devices

arXiv:2608.26204v1 Announce Type: cross Abstract: Computer Use Agents (CUAs) are increasingly deployed to navigate mobile and desktop applications on behalf of users, yet no benchmark comprehensively

After Meta avocado we get watermelon, due in November

Quote: ... developing a new A.I. model intended to be as powerful as Anthropic’s cutting-edge models. ... In July, while developing Watermelon, Meta paused and later resumed a stage of A.I. developmen

b10665

model-releasesllama-cpp-releases

model: add DSpark support for Nemotron3.5 (#27804) model: add DSpark support for Nemotron3.5 Update src/models/dflash.cpp Co-authored-by: Sigbjørn Skjæret sigbjorn.skjaeret@huggingface.co Co-authored-

Behavior2Trip: Towards Personalized Travel Planning via User Behavior Trajectory

arXiv:2608.26807v1 Announce Type: cross Abstract: Travel planning agents assist users in generating personalized travel plans by modeling their individual preferences. Existing agents either rely on e

Consolidating RLVR Capabilities Across Domains: A Deep Dive into Fusion Paradigms

arXiv:2608.27409v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) improves specific capabilities of large language models, but covering multiple capabilities often

Don't Overthink, Don't Underthink: Toward Adaptive Reasoning in Agentic AI

arXiv:2608.26442v1 Announce Type: new Abstract: Recent advances in Large Language Models (LLMs) have shown that increased inference-time reasoning can improve performance on complex tasks. However, ma

Evaluating human and LLM screening workflows in a conceptually complex scoping review: Recall--workload trade-offs and run-to-run consistency

arXiv:2608.26885v1 Announce Type: new Abstract: Background. Large language models (LLMs) are increasingly used for screening in evidence synthesis, where false negatives can remove relevant studies be

Invocation-Level Reliability of Tool-Using Agents

arXiv:2608.26189v1 Announce Type: new Abstract: Tool-using agents fail two ways: choosing the wrong tool, or forming wrong arguments, and an early failure of either kind can silently corrupt everythin

Knowledge-Verified Emergent Deception in LLM Agents Under Conflicting Incentives

arXiv:2608.26372v1 Announce Type: cross Abstract: Large language models are increasingly deployed as autonomous agents serving users on behalf of companies, placing them in settings where user and dep

Omni-Interactive Universal Embedder

arXiv:2608.27044v1 Announce Type: new Abstract: Multimodal representation learning has been shifting from traditional two-tower architectures to large language model (LLM)-based embedders due to their

PredVLA: A Sub-Million-Parameter Predictive-Coding Policy for Robot Manipulation

model-releasesarxiv-cs-ro

arXiv:2608.26673v1 Announce Type: new Abstract: Large pretrained vision-language-action models dominate modern robot-manipulation benchmarks, but it remains unclear how much model scale is necessary f

Qwen3.8-Flash-Next (UD-IQ4_XS) on 2x RTX 3060 + 7800X3D, from initial 36 tps prefill to 400 tps and other benchmarks (-sm tensor trap) + VRAM/RAM usage

TL;DR: llama.cpp with --load-mode mmap used 21-32 GB of RAM, with ik_llama.cpp using 106-108 GB. -sm tensor killed my prefill, changing to -sm layer went from 36 tps to 135 tps. After that, -ubatch 20

Reason in the Words You Speak: Idiolectal Paraphrasing Off-Policy Traces for Reasoning Distillation in VideoLLMs

arXiv:2608.26684v1 Announce Type: new Abstract: Recent large language models achieve strong performance on complex reasoning tasks, where reinforcement learning with Group Relative Policy Optimization

safetyarxiv-cs-cv

Rethinking Image Processing for the Age of AI: A Problem-First Framework for Scientific Progress

arXiv:2608.26833v1 Announce Type: new Abstract: Modern AI has greatly expanded the capabilities of image processing. However, the ready availability of powerful models, public datasets, and benchmark

Unifying Detection and Adaptation in Task-Free Continual Learning

arXiv:2608.27070v1 Announce Type: cross Abstract: To mitigate catastrophic forgetting in downstream continual learning (CL) for large language models (LLMs), existing methods typically constrain param

yall are sleeping on qwen 3.8 27b q2 + q2 dflash + q5 kv

ok bit more context: it's actually a QAT Q2 for Qwen 3.8 27 B: https://huggingface.co/sdkyuan/qwen3.8-27B-qat-q2_0-gguf QAT Q2 for DFlash model: https://huggingface.co/HermiHg/Qwen3.8-27B-DFlash2-Q2_K

A Layered Security Framework Against Prompt Injection in RAG-Based Chatbots

arXiv:2606.19660v2 Announce Type: replace-cross Abstract: Prompt injection is ranked as the most critical vulnerability in large language model (LLM) deployments by the OWASP Top 10 for LLM Applicatio

A Self-Evolving Multi-Agent Framework Defense against LLM Jailbreak Attacks

arXiv:2608.26008v1 Announce Type: cross Abstract: Large language models (LLMs) remain vulnerable to jailbreak attacks that exploit techniques such as role-playing, obfuscation, code transformation, an

AdaVDR: Adaptive Tool Use and Reflection for Video Deep Research

arXiv:2608.25559v1 Announce Type: new Abstract: Video deep research answers complex questions by jointly understanding video content and retrieving external knowledge from the open Web. However, diver

b10653

model-releasesllama-cpp-releases

llama: model_loader: add TENSOR_READ_LAZY (#27794) llama: model_loader: add TENSOR_GET_ROW_LAZY add --tensor-read-lazy rename to TENSOR_READ_LAZY gen docs address comments Website: https://llama.app A

Research

Beyond Pairwise Feedback: Listwise Vision-Language Supervision for Preference-Based Reward Learning

arXiv:2608.25350v1 Announce Type: new Abstract: Vision-language models (VLMs) have emerged as a powerful source of supervision for reinforcement learning, enabling agents to leverage rich semantic kno

researcharxiv-cs-lg

DCEO: Direct Causal Effect Optimization for Long-Term User Value Modeling in E-commerce Search

arXiv:2608.25635v1 Announce Type: new Abstract: Industrial e-commerce search systems ultimately aim to optimize the user-level long-term objective, such as n-day cumulative purchases or gross merchand

safetyarxiv-cs-lg

E2Rank: Unifying Text Embedding and Listwise Reranking for Effective and Efficient Search

arXiv:2510.22733v3 Announce Type: replace Abstract: Text embedding models deliver competitive retrieval performance with high efficiency, but their ranking fidelity remains limited compared to LLM-bas

Local LLM harness for reverse engineering software?

Does anyone know of a decent reverse engineering harness/workspace setup for local models that I can just point the model at and have it go to work until it's reversed most if not all of the functions

Mitigating LLM sycophancy with RL-based fine-tuning: Bayesian Truth Serum approach

arXiv:2608.25267v1 Announce Type: new Abstract: Large language models (LLMs) frequently exhibit sycophancy: they adapt their answers to a user's stated beliefs or preferences instead of reporting what