Guilherme Monteiro

guicybercode's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

commented a paper 15 minutes ago

Think Less, Locally: Local Preference Learning for Reasoning Compression

Paper • 2611.04814 • Published 8 days ago • 64 •

14

New activity in tardellirs/colibri-embed-ptbr about 16 hours ago

ONNX RAM peak vs fp32 on a 4 vCPU box

4

#1 opened about 23 hours ago by

guicybercode

commented a paper about 20 hours ago

Signed but Steerable: Measuring and Closing the Text-Vector Integrity Gap in Offline Retrieval-Augmented Generation

Paper • 2611.06498 • Published 6 days ago • 5 •

10

liked 2 models about 23 hours ago

FMendonca/qwen3-1.7b-agente-suporte-ptbr-gguf

Text Generation • 2B • Updated 2 days ago • 121 • 1

tardellirs/colibri-embed-ptbr

Sentence Similarity • 0.2B • Updated about 21 hours ago • 7.09k • • 107

upvoted 3 papers about 23 hours ago

Authorization Drift in LLM Agents: Evaluating Boundary Compliance in Multi-Step Tool Use

Paper • 2611.09113 • Published 4 days ago • 6

AutoCompress: Automated Hardware-Guided Model Compression

Paper • 2611.08290 • Published 5 days ago • 9

Signed but Steerable: Measuring and Closing the Text-Vector Integrity Gap in Offline Retrieval-Augmented Generation

Paper • 2611.06498 • Published 6 days ago • 5

commented 2 papers about 23 hours ago

AutoCompress: Automated Hardware-Guided Model Compression

Paper • 2611.08290 • Published 5 days ago • 9 •

3

Signed but Steerable: Measuring and Closing the Text-Vector Integrity Gap in Offline Retrieval-Augmented Generation

Paper • 2611.06498 • Published 6 days ago • 5 •

10

New activity in FMendonca/qwen3-1.7b-agente-suporte-ptbr-gguf about 23 hours ago

Template embutido no GGUF e mediana com 4 threads

#1 opened about 23 hours ago by

guicybercode

New activity in Horizon-Labs/prompt-injection-guard-base 3 days ago

Per-language false alarms and the int8 ONNX file on long inputs

5

#1 opened 19 days ago by

guicybercode

New activity in apollowmm/ragsec-prompt-injection 3 days ago

Ungrounded answers without an attacker: retrieval or model?

❤️ 1

2

#1 opened 6 days ago by

guicybercode

commented a paper 6 days ago

Think Less, Locally: Local Preference Learning for Reasoning Compression

Paper • 2611.04814 • Published 8 days ago • 64 •

14

New activity in RelyD/prompt-injection-detector 6 days ago

Multilingual-cased on Portuguese prompts?

2

#1 opened 7 days ago by

guicybercode

New activity in Horizon-Labs/prompt-injection-eval-suite 6 days ago

Native or translated rows in the synthetic multilingual configs

2

#2 opened 19 days ago by

guicybercode

commented 2 papers 6 days ago

You don't need long context

Paper • 2611.05306 • Published 8 days ago • 23 •

8

Think Less, Locally: Local Preference Learning for Reasoning Compression

Paper • 2611.04814 • Published 8 days ago • 64 •

14

commented a paper 7 days ago

Follow the Gradient: Tracing Safety Degradation Back to Training Data

Paper • 2610.37481 • Published 15 days ago • 32 •

3

New activity in Tobiashopf/ptbr-finance-eval 7 days ago

Casos de injeção indireta?

#2 opened 7 days ago by

guicybercode