RubricWorld improves LLM grading accuracy with world model-guided rubric optimization
Raises gpt-5-mini accuracy from 18.20% to 46.70% on ASAP2.0 exceeding baseline by 22.60 points.
40
AI model
GPT-5-mini
- Mentions · 7d0
- Mentions · 30d2
- All mentions7
- First seenAug 24, 2026
- Last mentioned10d ago
Timeline
7 articles · newest firstHumans Defer to Confident but Wrong LLMs
Participants adopted planted falsehoods from LLMs at 65% rate regardless of briefing falsification level, collapsing trust calibration.
55
Write Like It's 1866: LLMs Relearn Telegraphese
A benchmark finds LLMs can store records in telegraph-style cablese, cutting tokens by up to 49 percent without hurting later answers.
46
Threat-Preserving Representation Sensitivity in Agent-Security Benchmarks
Agent-security benchmark ASR scores shift up to 13 percentage points when tool names change, showing single-representation robustness scores may not generalize.
50
Knowledgator Releases GLiFormer: A 575M-Parameter Encoder That Hits 91.10 F1 on Nested JSON Extraction Without Generating Tokens
Knowledgator released GLiFormer, an Apache-2.0 encoder (264M/575M) handling NER, classification, relations, and nested JSON extraction, scoring 91.10 F1.
35
Geometry of Values: Task Vector Composition for Ethical Preference Alignment in Language Models
Researchers build a 12,000-dilemma multilingual dataset and use orthogonalized task vectors to isolate and steer ethical value preferences in LLMs.
30
What researchers learned about building an LLM security workflow
Oslo and FFI researchers show structured agentic workflows lift LLM alert-triage accuracy from 0% to about 93% on malicious cases.
55
Appears with
- Norwegian Defence Research Establishmentvendor · 1
- Gemma-4-31B-ITai model · 1
- Llama-3.2ai model · 1
- GLiFormer Large v1ai model · 1
- GLiFormer Base v1product · 1
- GLiFormer Base v1ai model · 1
- AIT Log Data Set V1.1product · 1
- Claude Haiku 4.5ai model · 1
- Suricataproduct · 1
- GLiNER-Relexai model · 1
- GLiFormer Large v1product · 1
- gemma-4-26bai model · 1
- GLiFormerproduct · 1
- GLM-5.3-Flashai model · 1
- Gemma 3:27Bai model · 1
- qwen3.8-27bai model · 1
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.