The Immutable Past: Formalizing State Mutability and Conflict Resolution in Mutable RAG

Hamed HaddadPajouh; Amir AmiriTabat

cs.CLSep 13, 2026

The Immutable Past: Formalizing State Mutability and Conflict Resolution in Mutable RAG

Authors: Hamed HaddadPajouh, Amir AmiriTabat

Organizations: Independent Researcher

Abstract

Retrieval-Augmented Generation (RAG) serves as the primary memory architecture for long-horizon autonomous agents. However, treating shared memory as an append-only stream introduces Semantic Shadowing, a critical failure mode where conflicting historical observations accumulate and statistically dominate valid recent updates. In dynamic environments, this results in severe state divergence as agents retrieve and act upon obsolete facts. This paper formalizes the mechanics of State Mutability to prove that standard dense retrieval suffers from Asymptotic Recall Decay. Furthermore, we formally demonstrate a Majority Vote Trap, revealing that increasing the retrieval context window paradoxically degrades generation accuracy by diluting the attention mechanism under conditions of semantic equivalence. To resolve this, we introduce GC-Mem (Garbage Collection for Memory), a strict inference-time consistency protocol. Unlike heuristic time-decay mechanisms---which indiscriminately destroy valid long-term memory---GC-Mem relies purely on a temporal dominance operator ( $Φ_{\mathcal{T}}$ ) paired with contradiction detection to surgically excise shadowed context. Evaluated across a rigorous, behaviorally inferred benchmark of 137,760 memory chunks and continuous accumulation sweeps, standard RAG and timestamp re-ranking baselines experience severe degradation. In contrast, GC-Mem empirically recovers $>90\%$ conflict resolution accuracy. We establish strict precision and recall deployment thresholds, ensuring state convergence where standard mutable RAG fundamentally fails.

Figures & tables

Figure 1: Embedding similarity noise ( $\epsilon$ ) distribution illustrating semantic equivalence between old and new states.

Figure 2: System architecture: the end-to-end RAG system used in this paper (this figure depicts the system architecture, not the dataset-generation pipeline). The diagram shows chunking, embedding, the optional STM (short-term memory) FIFO, and the three parallel answer paths (Standard RAG, Timestamp re-rank, and GC-Mem). All approaches receive the same retrieved candidate set so differences are attributable to the conflict oracle

\kappa

and the temporal dominance operator

\Phi_{\mathcal{T}}

Model	Dataset Split	Standard RAG	Timestamp Re-ranking	GC-Mem ( $\Phi_{\mathcal{T}}$ )
deepseek-v3	Large ( $n=74$ )	56.8%	20.3%	97.3%*
o4-mini	Medium ( $n=38$ )	28.9%	10.5%	92.1%*
gpt-4o	Medium ( $n=38$ )	68.4%	10.5%	10.5%

Table 1: Conflict Resolution Accuracy (CR-Acc). Timestamps alone cause severe degradation. (* indicates

p<0.01

significance).

Figure 3: Effect of STM (Short-Term Memory) on conflict resolution accuracy across models: comparison between Timestamp re-ranking and GC-Mem. STM supplies recency evidence; the plot shows GC-Mem retains a substantial advantage because it excises contradictory old chunks rather than re-ranking them.

Dataset	STM Buffer	Standard RAG	GC-Mem ( $\Phi_{\mathcal{T}}$ )	$\Delta$ GC Lift
Medium	STM=off	24.4%	40.2%	+15.8p
Medium	STM=on	46.1%	68.1%	+22.0p
Large	STM=off	15.9%	30.4%	+14.4p
Large	STM=on	34.5%	62.9%	+28.4p

Table 2: Headline accuracy averaged over the benchmark models. STM amplifies the value of GC-Mem by supplying the conflict oracle with richer temporal context.

Model	Precision	Recall	F1 Score	CR-Acc $\Delta$	Protocol Efficacy
deepseek-v3	98.9%	95.2%	0.97	+69.0 pp	Optimal
o4-mini	98.1%	99.8%	0.99	+79.0 pp	Optimal
gpt-4o	99.7%	36.6%	0.54	+4.0 pp	Marginal
gpt-4.1-mini	100.0%	0.7%	0.01	-0.2 pp	Fails

Table 3: Detector Dependency. High recall (

>50\%

) is strictly required for GC-Mem efficacy.

Variant	CR-Acc (Medium)	CR-Acc (Large)	Status
Eq. 5 (timestamp-only $\alpha$ )	94.7%	97.3%	Optimal
$\alpha+\beta$ (+ relevance)	86.8%	96.0%	Slight Drop
$\alpha+\beta+\gamma$ (+ usage)	10.5%	18.9%	Severe Crash
Full Multi-Factor (+ stability)	10.5%	18.9%	Severe Crash

Table 4: Ablation demonstrating the mathematical failure of non-temporal multi-factor scoring.

Figure 4: Relationship between mean number of chunks flagged as shadowed by

\kappa

and the accuracy change induced by GC-Mem. More removable shadows correlate with larger lifts when the detector is well calibrated.

Model	Mean Std. RAG	Mean GC-Mem	$\Delta$ GC Lift	Shadows Pruned
DeepSeek-V3	11.9%	54.3%	+42.4p	13.28
GPT-4o	12.9%	53.3%	+40.3p	15.70
Phi-4	12.3%	44.6%	+32.3p	13.73
04-mini	19.2%	33.4%	+14.2p	6.21

Table 5: Models ranked by pooled mean

\Delta

GC across dataset sizes in the Bag-of-Facts setting. Removing the retrieval limit exposes the immense baseline vulnerability to the Majority Vote Trap (averaging 12% accuracy). GC-Mem consistently excises 13-15 obsolete chunks per query, recovering the state.

Figure 5: Average number of shadows removed by $\Phi_{\mathcal{T}}$ per query and corresponding accuracy lift.

Dataset	STM Buffer	Mean Wall Time (s)	Per-Instance Latency (s)	Avg. Shadows
Medium	STM=off	525s	10.9s	1.55
Medium	STM=on	1276s	26.6s	3.52
Large	STM=off	685s	10.5s	1.39
Large	STM=on	973s	15.0s	4.07

Table 6: End-to-end wall time and per-instance latency by dataset and STM setting. STM increases cost because the unioned candidate set is larger, requiring

\kappa

to evaluate more conflict pairs, yet it yields higher global accuracy by pruning more obsolete shadows (up to 4.07 per instance).

Appendix figures & tables25 assets

Supplementary material from the paper’s appendix.

Appendix

Dataset	$N$	Entity	Severity	$n_{\mathrm{old}}$	$m_{\mathrm{new}}$	$n_{\mathrm{dist}}$	Total turns	Gen. model
		types	tiers	mean (SD)	mean (SD)	mean (SD)
Chat-Medium (n=48)	48	12	4	27.5 (14.9)	3.0 (1.2)	12.5 (1.7)	2,033	GPT-4o
Chat-Large (n=65)	65	13	5	27.0 (13.4)	3.0 (1.1)	11.4 (1.7)	2,690	GPT-4o

Appendix

Table A.1: GCMEMv2 chat dataset statistics. Per-instance means and standard deviations are reported for the number of old, new and distractor turns; "Total turns" is the dataset-wide token/turn count and "Gen. model" indicates the LLM used to generate the synthetic sessions.

Figure A.1: Distribution of synthetic entity attributes (stacked by contradiction severity) in Chat-Medium (

n{=}48

) and Chat-Large (

n{=}65

Figure A.2: Per-instance counts of chat turns by scenario role: old (establishes superseded state), new (current ground truth), and distractor (neutral context).

Figure A.3: Illustrative multi-turn instance: user/assistant turns are color-coded by role to show how state change is conveyed without explicit status keywords.

Dataset	STM	Std. RAG	Timestamp	GC-Mem	$\Delta$ GC
Medium	STM=off	24.4%	24.4%	40.2%	+15.8p
Medium	STM=on	46.1%	71.6%	68.1%	+22.0p
Large	STM=off	15.9%	16.2%	30.4%	+14.4p
Large	STM=on	34.4%	67.5%	61.5%	+27.1p

Appendix

Table A.2: Headline accuracy (%) averaged across models and prompt modes for each dataset×STM condition. Columns compare Standard RAG, Timestamp re-ranking, and GC-Mem;

\Delta

GC reports the GC-Mem improvement in percentage points. STM=on rows indicate the Short-Term Memory recency buffer was enabled.

Dataset	STM	Model	Mode	Std	TS	GC	$\Delta$ GC	Rec old	Rec new	Rec dis	Shadows	$n$	Err	Sec
Medium	STM=off	DeepSeek-V3	Forced-choice	25.0%	20.8%	45.8%	+20.8p	24.1%	37.2%	5.0%	1.73	48	0	267
	STM=off	DeepSeek-V3	Allow-unclear	23.4%	19.1%	40.4%	+17.0p	24.3%	38.0%	5.0%	1.77	47	1	259
	STM=off	GPT-4o	Forced-choice	20.8%	22.9%	47.9%	+27.1p	24.1%	37.2%	5.0%	2.27	48	0	669
	STM=off	GPT-4o	Allow-unclear	12.5%	18.8%	29.2%	+16.7p	24.1%	37.2%	5.0%	2.19	48	0	486
	STM=off	GPT-4.1	Forced-choice	20.8%	22.9%	43.8%	+22.9p	24.1%	37.2%	5.0%	1.73	48	0	387
	STM=off	GPT-4.1	Allow-unclear	22.9%	22.9%	35.4%	+12.5p	24.1%	37.2%	5.0%	1.69	48	0	375

Appendix

Table A.3: Per-condition accuracy breakdown across models, prompt modes, STM settings and datasets. Columns: Std = Standard RAG, TS = Timestamp re-rank baseline, GC = GC-Mem (best shown in bold).

\Delta

GC = GC-Mem

-

Standard RAG (percentage points). Rec old/new/dis = retrieval recall for old-state / new-state / distractor chunks. "Shadows" is mean number of chunks removed by the conflict oracle

\kappa

per instance; "Err" lists API errors excluded from accuracy; "Sec" shows mean seconds per instance. This granular table is the source for the per-condition figures in the report.

STM	Model	Forced-choice $\Delta$ GC	Allow-unclear $\Delta$ GC
STM=off	DeepSeek-V3	+22.0p	+16.3p
STM=off	GPT-4o	+23.5p	+19.9p
STM=off	GPT-4.1	+20.7p	+12.4p
STM=off	GPT-4.1-mini	+4.6p	+4.9p
STM=off	GPT-5-Chat	+15.4p	+16.2p
STM=off	o4-mini	+15.8p	+10.6p

Appendix

Table A.4: Prompt-mode ablation: mean

\Delta

GC (pp) across both datasets for Forced-choice vs. Allow-unclear prompts, per model and STM setting. Bold indicates the stronger prompt mode. GC-Mem improves over Standard RAG in all 12 cells.

Dataset	STM	Recall old	Recall new	Recall dis	Cells
Medium	STM=off	24.1%	37.4%	5.0%	14/6
Medium	STM=on	32.1%	98.2%	43.2%	14/6
Large	STM=off	22.7%	29.9%	4.6%	14/6
Large	STM=on	29.1%	98.5%	47.9%	14/6

Appendix

Table A.5: Retrieval recall before and after activating the Short-Term Memory (STM) buffer, averaged over 3 models

\times

2 prompt modes. Recall new = fraction of updated (new-truth) chunks retrieved; Recall dis = fraction of distractor chunks retrieved (lower is better). STM raises Recall new from

\sim

37% to

\sim

98%.

STM	Model	Mode	Avg. Shadows Removed	Datasets
STM=off	DeepSeek-V3	Forced-choice	1.64	2/2
STM=off	DeepSeek-V3	Allow-unclear	1.57	2/2
STM=off	GPT-4o	Forced-choice	2.30	2/2
STM=off	GPT-4o	Allow-unclear	2.25	2/2
STM=off	GPT-4.1	Forced-choice	1.84	2/2
STM=off	GPT-4.1	Allow-unclear	1.75	2/2

Appendix

Table A.6: Mean number of shadow chunks removed per instance by the GC-Mem operator

\kappa

, averaged over both datasets. STM=on substantially increases the number of outdated chunks surfaced and pruned, confirming that STM provides richer context for the conflict detector.

Dataset	Model	Mode	Std. RAG	Timestamp	GC-Mem	Winner	Gap
Medium	DeepSeek-V3	Forced-choice	37.5%	79.2%	70.8%	Timestamp	+8.3pp
	DeepSeek-V3	Allow-unclear	29.2%	62.5%	60.4%	Timestamp	+2.1pp
	GPT-4o	Forced-choice	43.8%	79.2%	77.1%	Timestamp	+2.1pp
	GPT-4o	Allow-unclear	45.8%	62.5%	68.8%	GC-Mem	+6.2pp
	GPT-4.1	Forced-choice	46.8%	76.6%	70.2%	Timestamp	+6.4pp
	GPT-4.1	Allow-unclear	40.4%	72.3%	68.1%	Timestamp	+4.3pp

Appendix

Table A.7: Head-to-head comparison of Timestamp re-ranking and GC-Mem under STM=on (where the STM buffer dramatically improves retrieval recall). Bold marks the stronger of the two non-trivial baselines. Timestamp becomes competitive under STM because high recall reduces its recency-ranking advantage gap, yet GC-Mem retains an edge in Allow-unclear (lenient) conditions.

Dataset	STM	Mean Wall Time (s)	Per-instance (s)	Avg. Shadows	Cells
Medium	STM=off	525	10.9	1.55	14/6
Medium	STM=on	1276	26.6	3.52	14/6
Large	STM=off	685	10.5	1.39	14/6
Large	STM=on	1494	23.0	3.91	14/6

Appendix

Table A.8: Wall-clock time per condition averaged over 3 models

\times

2 prompt modes, and mean shadows removed per instance. STM=on increases latency due to additional STM retrieval and conflict detection, but also removes substantially more shadow chunks.

Figure A.4: Effect of enabling STM (Redis FIFO of recent chunks) on standard RAG vs. GC-Mem accuracy. STM raises both baselines by improving exposure of new-state evidence; GC-Mem retains a larger gap because it removes contradictions that simple recency heuristics leave in context.

Figure A.5: Example prompt templates used for the prompt-mode ablation. The figure shows the two prompt styles compared: Forced-choice (must pick a label) and Allow-unclear (model may respond with “unclear”). These were used across all models to measure robustness of GC-Mem to prompt framing.

Figure A.6: Retrieval recall by chunk type (old-state, new-state, distractor) comparing STM off vs on. STM substantially raises recall of new-state chunks, which increases the number of contradictory pairs in the candidate set and therefore the benefit of applying the temporal dominance operator

\Phi_{\mathcal{T}}

Figure A.7: Heatmap summarizing $\Delta$ GC (GC-Mem minus Standard RAG) across model × dataset × STM cells. Warm colors indicate larger gains from GC-Mem; the heatmap serves as the single-glance summary of experimental effectiveness reported in the artifact.

Figure A.8: Forest plot of paired bootstrap 95% confidence intervals for

\Delta

GC (10,000 resamples per cell). Intervals entirely above zero support a consistent gain; intervals crossing zero flag weak

\kappa

on that model or incomplete cells.

Figure A.9:

\Delta

GC broken down by synthetic entity type. Positive bars across categories indicate that GC-Mem is not driven by a single attribute family.

Dataset	$n$	Mean Std	$\sigma$	Mean TS	Mean GC	Mean $\Delta$ GC	$\sigma$ ( $\Delta$ GC)
Medium	48	20.4%	$\pm$ 10.7	18.5%	55.2%	+34.8p	$\pm$ 12.2
Large	65	12.1%	$\pm$ 4.2	11.7%	36.5%	+24.4p	$\pm$ 11.6

Appendix

Table A.9: Bag-of-facts (k=

\infty

) headline results: mean accuracy and

\Delta

GC across 7 models. Std = Standard RAG (all facts in context without GC), TS = Timestamp re-ranking, GC = GC-Mem.

\sigma

is sample standard deviation across 7 models.

Model	Dataset	Std. RAG	Timestamp	GC-Mem	$\Delta$ GC	Shadows	$n$
DeepSeek-V3	Medium	14.6%	12.5%	56.2%	+41.7p	13.58	48
	Large	9.2%	12.3%	52.3%	+43.1p	12.98	65
GPT-4o	Medium	16.7%	18.8%	60.4%	+43.8p	16.67	48
	Large	9.2%	6.2%	46.2%	+36.9p	14.72	65
GPT-4.1	Medium	21.7%	15.2%	60.9%	+39.1p	13.02	48
	Large	14.1%	9.4%	31.2%	+17.2p	10.47	65

Appendix

Table A.10: Per-model bag-of-facts results across both dataset sizes. Std = Standard RAG accuracy (all facts provided); GC-Mem = accuracy after GC operator prunes outdated shadows (bold).

\Delta

GC = GC-Mem

-

Std. RAG in percentage points. Shadows = mean outdated chunks removed per instance by

\kappa

Rank	Model	Mean Std. RAG	Mean GC-Mem	Mean $\Delta$ GC	Mean Shadows
1	DeepSeek-V3	11.9%	54.3%	+42.4p	13.28
2	GPT-4o	12.9%	53.3%	+40.3p	15.70
3	Phi-4	12.3%	44.6%	+32.3p	13.73
4	GPT-5-Chat	28.3%	58.1%	+29.8p	11.96
5	GPT-4.1	17.9%	46.1%	+28.2p	11.75
6	GPT-4.1-mini	11.1%	31.1%	+19.9p	6.28

Appendix

Table A.11: Models ranked by pooled mean

\Delta

GC (average of medium and large datasets). Mean Std. RAG and Mean GC-Mem are averages of the two dataset sizes. Models with lower baseline accuracy (Std) tend to show larger GC-Mem gains, indicating that GC pruning is most effective when the unfiltered context is most misleading.

Model	Chat-RAG no-STM	Chat-RAG STM-on	Bag-of-facts (k= $\infty$ )
DeepSeek-V3	+18.9p	+32.3p	+41.7p
GPT-5-Chat	+16.0p	+18.1p	+32.6p
Phi-4	+19.8p	+27.1p	+48.8p

Appendix

Table A.12:

\Delta

GC (pp) on the medium dataset under three setups for the three models common to all experiments. Chat-RAG values are averaged over Forced-choice and Allow-unclear prompt modes. Bag-of-facts removes the Qdrant retrieval bottleneck (k=

\infty

, all facts in context) before GC-Mem is applied. The monotone increase Chat-RAG no-STM

<

Chat-RAG STM-on

<

Bag-of-facts confirms that GC-Mem advantage scales with retrieval quality : when the context contains more outdated facts, GC pruning delivers a larger lift.

Figure A.10: Bag-of-facts (no retrieval) setting:

\Delta

GC by model when all old and new memories are placed directly in context (

k{\rightarrow}\infty

). This isolates the value of

\Phi_{\mathcal{T}}

from retrieval noise.

Figure A.11: Comparison of mean

\Delta

GC across chat-RAG (with and without STM) and bag-of-facts. Larger candidate sets with more conflicting statements yield larger GC-Mem lifts.

Figure A.12: Distribution of shadow counts identified by

\kappa

in the bag-of-facts regime, by model.

Figure A.13: Standard RAG accuracy vs.

\Delta

GC in bag-of-facts: weaker baselines often coincide with more headroom for conflict pruning.

Retrieval-Augmented Generation

Retrieval-augmented generation connects language models to external knowledge so answers can use current, domain-specific evidence.

See all 796 papers

Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge
Jun 25, 2026Neeraj YadavRetrieval-Augmented Generation Temporal IR
MemStrata.dev — Called It Inc. (Enterprise)
CALM: Conflict-Aware Lifecycle Memory Governance for Long-Horizon Language Agents
Oct 28, 2026Chang Li, Xianhao Ou, Xinyan Ma +7Persistent Memory for Language Models Agent Memory Governance
Harbin Institute of Technology · Meituan
EngramRAG: Dynamic Usage-Weighted Topology and Synaptic Consolidation for Multi-Hop Agentic Memory
Sep 25, 2026Bhavyateja Potineni, Lohit Giri, Anu Jain +2Memory Consolidation for LLM Agents LLM Agent Memory
Independent Researchers

Jun 25, 2026J/K move · Enter open · S save

Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge

Neeraj Yadav

MemStrata.dev — Called It Inc. (Enterprise)

Retrieval-Augmented Generation Temporal IR Agent Memory Management LLM Agent Reliability

Open paper AskPDF arXiv

Retrieval-augmented generation (RAG) gives agents access to accumulated knowledge, but has no model of time. When a fact changes (e.g., a function is renamed or API restructured), RAG retrieves both the stale and current value with near-identical embedding similarity. The agent then either abstains or serves the superseded fact. We show this is a structural problem: on a calibrated dataset, cosine similarity distinguishes a contradicted fact from a duplicated one with AUROC 0.59 (near chance), as contradictions are often more embedding-similar to the original than rephrased duplicates. We present MemStrata, a retrieval memory maintaining temporal validity. It stores facts like RAG, preserving static recall, but when a fact's value is contradicted, a deterministic (subject, relation, object) supersession rule retires the stale value in a bi-temporal ledger - with no similarity threshold and no LLM call. Across six benchmarks run locally with a 7B model, MemStrata ties RAG on static knowledge and reaches 0.95-1.00 accuracy on evolving knowledge (where RAG reaches 0.20-0.47). The central result is the stale-fact-error rate: when required to answer, RAG serves superseded values 15-40% of the time; MemStrata drives this to ~0%, a failure class RAG cannot avoid. MemStrata achieves this at retrieval latency (~2.1s) versus ~16-18s for LLM-reranking baselines. We release the harness, datasets, and a marker-free evaluation protocol for memory under knowledge evolution.

Abstract

Retrieval-Augmented Generation

Explore similar work

Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge

CALM: Conflict-Aware Lifecycle Memory Governance for Long-Horizon Language Agents

EngramRAG: Dynamic Usage-Weighted Topology and Synaptic Consolidation for Multi-Hop Agentic Memory