multi-year-authentic-personal-photo-video-archives-turned-into-memory CLM-0455On multi-year authentic personal photo/video archives turned into memory streams, frontier multimodal models and structured memory systems recall facts and infer persona far better than they predict a user's future choices: predicting or recommending in novel contexts caps around 59 on a 0-100 scale for every system tested, and compressing the history ~70x into a memory system costs nothing on that tier while costing a lot on factual recall, so predictive personalization is limited by preference representation rather than by retrieval or context length.
Capability: Remembering across sessions
Observed on
ReaLMem: 2,508 sessions from 7 participants over 6+ years, 1,629 QA pairs; readers Gemini-3.0-Flash, GPT-4.1-mini, GPT-5.4-mini; memory systems Mem0, MemOS, EverMemOS on a shared G.
Model cards: GPT-4
Sources
- Validated by: Swapped into EverMemOS's pipeline with retrieval and backbone fixed: beat full context on persona inference for all three readers (+0.6/+4.0/+4.3) and matched or beat it on prediction (0.0/+3.5/+1.8) at ~6k tokens vs ~123k; ablating the stability weighting lowered persona inference for all readers but raised open-ended assistance by about 2.4 points. Measured on one benchmark with 7 participants; [truncated]
Disagreeing is the most useful thing you can do here. Nothing on this page waits for a reviewer: the counts above move when evidence lands, and every source stays attached with who brought it.
Related claims
- When the preference needed to answer a query lies in a different concept space from the cues in a user's dialogue history (e.g. inferring an emotion-regulation strategy from personality cues), retrieval-based and profile-summarization personalization methods barely beat a no-history baseline, and dumping the full history into the prompt does worse than no history at all.Remembering across sessions · reviewed by AI
- For persistent-memory agents, any memory-write rule that decides using only recency and provenance is stuck on a single tradeoff — accepting more genuine preference updates means admitting more poisoned ones — because an adversary who can launder a claim through the user's own channel matches the statistics a genuine revision produces; conditioning the write decision on the inferred authenticity of the claim given the interaction history moves both axes at once.Remembering across sessions · reviewed by AI
- For streaming video question answering with a frozen video-LLM, appending retrieved historical frames as extra visual context helps less than optimizing a small set of latent memory tokens at test time and then dropping the retrieved tokens before decoding; the gain is largest on questions requiring backward tracing over past history.Reasoning about time in video · reviewed by AI
- In retrieval-augmented personalization with a frozen 8-9B generator, prepending more retrieved user-history records does not monotonically improve quality — on the six LaMP tasks a fixed top-k profile peaked at k=5 and lost 0.8 quality points at k=10 while the share of cases made worse than an empty profile rose from 23.2% to 31.7% — so a learned per-request stopping decision beats any single fixed k.Remembering across sessions · reviewed by AI
- For long cooking videos where a question depends on ingredient changes no longer visible, vision-language models fail to carry observed events forward into later answers, and adding generic retrieved-context RAG makes it worse, while splitting memory into a temporally anchored event log plus an explicitly updated inventory ledger recovers much of the gap.Tracking state through a long task · reviewed by AI
Notes
Drafted from the paper by a model and filed unreviewed. Visible here so it can be read, not because anyone has vouched for it: it does not move any technique's standing and does not count toward the internal scorecard. Drafted confidence: medium. Falsifier as drafted: A system that raises predictive-personalization scores well above the ~59 band by adding retrieval depth or full raw context, or evidence that memory compression degrades the predictive tier as much as it degrades factual recall.