Skip to content
LMC Feed-Models, Papers, Benchmarks. Zero Fluff.Live

LMC Feed

LLM & AI News·RSS·Archive
|
1.OpenAI's GPT-5.6 Sol autonomously post-trained the smaller Luna model with a "fairly underspecified prompt"
74 pts·The Decoder··release
2.Please don't discontinue Gemini 2.5 Flash
57 pts·discuss.ai.google.dev··release
3.GPT-5.6 Sol Ultra produces proof of the Cycle Double Cover Conjecture [pdf]
50 pts·cdn.openai.com··release
4.Is Grok 4.5 Really More Token Efficient Than Claude Opus 4.8? I Checked the Numbers
68 pts·Towards AI··release
5.OpenAI staffer maps out which of GPT-5.6 Sol's five reasoning levels fits which task complexity
88 pts·The Decoder··release
6.OpenAI kills its Atlas browser after just eight months and folds everything into ChatGPT
38 pts·The Decoder··tools
7.Do You Need a Frontier Model as a Citation Verifier? Benchmarking Rubric LLMs for Deep-Research Source Attribution
42 pts·arXiv cs.CL··research
8.OpenAI says GPT 5.6 is the ‘preferred model’ for Microsoft Copilot 365 amid breakup chatter
53 pts·TechCrunch··release
9.KronQ: LLM Quantization via Kronecker-Factored Hessian
76 pts·arXiv cs.LG··research
10.What LLM Forecasters Know but Don't Say: Probing Internal Representations for Calibration and Faithfulness
47 pts·arXiv cs.CL··research
11.PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents
59 pts·arXiv cs.CL··research
12.ReCoLoRA: Spectrum-Aware Recursive Consolidation for Continual LLM Fine-Tuning
64 pts·arXiv cs.LG··research
13.Super Weights in LLMs and the Failure of Selective Training
52 pts·arXiv cs.LG··research
14.Can We Trust LLM's Logic? Quantifying Uncertainty, Coherence, and Robustness via a Graph-Based Framework
65 pts·arXiv cs.CL··research
15.Token-Flow Firewall: Semantic Runtime Auditing for Persistent AI Agents
74 pts·arXiv cs.CL··research
16.LLM-powered reasoning in agent-based modeling
70 pts·arXiv cs.AI··research
17.Evaluating SageMath-Augmented LLM Agents for Computational and Experimental Mathematics
89 pts·arXiv cs.AI··research
18.Scalable and Culturally Specific Stereotype Dataset Construction via Human-LLM Collaboration
75 pts·arXiv cs.CL··research
19.When Implausible Tokens Get Reinforced: Tail-Aware Credit Calibration for LLM Reinforcement Learning
81 pts·arXiv cs.CL··research
20.From Execution to Education: A Bloom-Aligned Framework for Measuring Educational Control in LLMs
55 pts·arXiv cs.CL··research
21.Holographic Neural PCFG for Unsupervised Parsing
32 pts·arXiv cs.CL··research
22.ICDAR 2026 HIPE-OCRepair Competition on LLM-Assisted OCR Post-Correction for Historical Documents
83 pts·arXiv cs.CL··research
23.SQuaD-SQL: Efficient Text-to-SQL with Small Language Models via LLM-Guided Knowledge Distillation
74 pts·arXiv cs.CL··research
24.Diarization-Guided Qwen-ASR Adaptation for Multilingual Two-Speaker Conversational Speech
85 pts·arXiv cs.CL··research
25.Cross-seed explainability using Procrustes-conditioned Joint End-to-end Top-K Sparse Autoencoders
58 pts·arXiv cs.CL··research
26.When the Judge Changes, So Does the Measurement: Auditing LLM-as-Judge Reliability
88 pts·arXiv cs.CL··research
27.Validity of LLMs as data annotators: AMALIA on authority
74 pts·arXiv cs.CL··research
28.Validating LLMs in social science: Epistemic threats and emerging norms
67 pts·arXiv cs.CL··research
29.From Prompts to Contracts: Harness Engineering for Auditable Enterprise LLM Agents
33 pts·arXiv cs.CL··research
30.How to Leverage Synthetic Speech for LLM-Based ASR Systems?
64 pts·arXiv cs.CL··research
30 of 500
LMC Feed - LLM & AI News (July 2026) | LM Market Cap