IBM Research post introduces Palette for design-language decks
The Hugging Face post says Palette teaches a small model to build decks in a chosen design language.
Stories per day · 30 days
1 RSS
Relatedaction-graphs1agentic-ai1ai-for-security1guardrails1huggingface1ibm-research1indirect-prompt-injection1llm-security1palette1prompt-injection1routing1security-operations1slide-generation1soc1tool-use-agents1
11 stories, newest first · one card per story
Stories tagged #small-models
LanG 2.0 enables energy-efficient agentic AI for security operations
LanG 2.0 improves campaign correlation to 51.81% and uses small local models at 650 times less energy than Claude Opus 5.5.
SpanGuard splits localization from authorization for small-model IPI defense
Authors report a 0.32% attack success rate and little loss of benign utility while using substantially smaller guardrail models.
NIOCIS picks tool-use agent next hops without generating text
On 3,875 held-out security-operations decisions, a typed encoder matched the logged hop 32.2% of the time in about 10 ms.
Continually Post-training Small Models via Synthetic Self-distillation
SynthSelf lets small models continually post-train by outsourcing in-distribution rewrites to a slightly larger same-family aligner.
SecJev: Bringing Security Expertise to System One Decision Models
SecJev releases 0.8B-9B security-specialized decision models; the 0.8B variant beats general Kev-9B by 20.51 points across 14 security decision tasks.
Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents
SSR selects pre-specified reasoning traces for small multimodal agents, cutting per-turn latency over 90%.
Explore Broadly, Reason Sharply: Push Small Models toward the Frontier via Sampling
Parallel Power Tempering improves inference-time reasoning, matching frontier models without RL post-training.
Mizar: A 159M-Parameter Audio-Language Model for Audio Understanding
Mizar, a 159M audio-language model, beats prior sub-200M models on MMAU, MMAR, and ADQA.
Coding Is Over. Get over It
A JPMorgan Chase engineer reflects on AI agents outpacing hand-coding, questioning ROI while predicting inference costs will become negligible.
Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO Steps
Hugging Face guide fine-tunes a 350M-parameter model with 100 GRPO steps to improve structured output reliability.