gemma-3-270m-it news

Words Speak Louder Than Order: A Behavioral Evaluation of Gemma 4

arXiv:2609.30716v1 Announce Type: cross Abstract: When a language model receives two conflicting documents as input, how does it decide which one to prioritize? Does it rely on how the sources are framed or the presentation order of the documents? We evaluated this behavior on Google's pre-trained G

Recurrence Is Not Enough: Causally Validating Multilingual SAE Translation Features in Gemma 2 and 3

arXiv:2609.04808v1 Announce Type: cross Abstract: Sparse autoencoder (SAE) features are increasingly used to explain and steer language-model behavior, but it remains unclear whether a feature found in one language context plays the same causal role when processing prompts in another language. We st

Gemma 4 Technical Report

arXiv:2607.02770v2 Announce Type: replace Abstract: We introduce Gemma 4, a new generation of open-weight, natively multimodal language models in the Gemma model family. Designed to advance compute efficiency and reasoning, the Gemma 4 model suite features dense and Mixture-of-Experts architectures,

Do Active SAE Feature Planes Carry More Holonomy? A Preregistered Reversal in Gemma

arXiv:2607.20522v1 Announce Type: new Abstract: This paper tests whether holonomy concentrates on active sparse-autoencoder (SAE) feature planes in Gemma 2 2B, a concrete operationalization of the broader semantic-concentration prediction. Holonomy is measured at the final-token layer-12 to layer-13

Hugging Face and Cerebras bring Gemma 4 to real-time voice AI

Fine-Tuning and Serving Gemma 4 31B on Google Cloud TPU: A Technical Comparison with GPU Baselines

arXiv:2605.25645v2 Announce Type: replace-cross Abstract: We present the first end-to-end demonstration of fine-tuning and serving Google's Gemma 4 31B model on TPU hardware, providing an empirical comparison of TPU and GPU platforms for large language model adaptation. Using LoRA on a Google TPU v5

Borrowed Geometry: Cross-Distribution Head-Importance Fingerprints of Frozen Pretrained Gemma 4 31B

arXiv:2605.00333v2 Announce Type: replace-cross Abstract: Frozen Gemma 4 31B weights pretrained exclusively on text, unmodified, transfer through a thin trainable interface to non-text modalities the substrate has never processed. On the L24--L29 slice (192 attention heads), an English-text TxtCopy

PSK at SemEval-2026 Task 9: Multilingual Polarization Detection Using Ensemble Gemma Models with Synthetic Data Augmentation

arXiv:2605.05159v1 Announce Type: new Abstract: We present our system for SemEval-2026 Task 9: Multilingual Polarization Detection, a binary classification task spanning 22 languages. Our approach fine-tunes separate Gemma~3 models (12B and 27B parameters) per language using Low-Rank Adaptation (LoR

Distilling Self-Consistency into Verbal Confidence: A Pre-Registered Negative Result and Post-Hoc Rescue on Gemma 3 4B

arXiv:2604.24070v1 Announce Type: cross Abstract: Small instruct-tuned LLMs produce degenerate verbal confidence under minimal elicitation: ceiling rates above 95%, near-chance Type-2 AUROC, and Invalid validity profiles. We test whether confidence-conditioned supervised fine-tuning (CSFT) with self

Gemma 4 VLA Demo on Jetson Orin Nano Super

Enabling Intrinsic Reasoning over Dense Geospatial Embeddings with DFR-Gemma

arXiv:2604.07490v1 Announce Type: new Abstract: Representation learning for geospatial and spatio-temporal data plays a critical role in enabling general-purpose geospatial intelligence. Recent geospatial foundation models, such as the Population Dynamics Foundation Model (PDFM), encode complex popu

Gemma 4, Phi-4, and Qwen3: Accuracy-Efficiency Tradeoffs in Dense and MoE Reasoning Language Models

arXiv:2604.07035v1 Announce Type: new Abstract: Mixture-of-experts (MoE) language models are often expected to offer better quality-efficiency tradeoffs than dense models because only a subset of parameters is activated per token, but the practical value of that advantage depends on end-to-end behav

Welcome Gemma 4: Frontier multimodal intelligence on device

Gemma 3n fully available in the open-source ecosystem!

Welcome Gemma 3: Google's all new multimodal, multilingual, long context open LLM

Google releases Gemma 2 2B, ShieldGemma and Gemma Scope

Welcome Gemma 2 - Google’s new open LLM

Fine-Tuning Gemma Models in Hugging Face

Welcome Gemma - Google’s new open LLM